From 6628f663b636675bcaea316f2deaddf337de480e Mon Sep 17 00:00:00 2001
From: baoshiwei <baoshiwei@shlanbao.cn>
Date: 星期五, 13 三月 2026 10:23:31 +0800
Subject: [PATCH] feat(米重分析): 新增稳态识别和预测功能页面并优化现有模型
---
app/pages/metered_weight_forecast.py | 716 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 716 insertions(+), 0 deletions(-)
diff --git a/app/pages/metered_weight_forecast.py b/app/pages/metered_weight_forecast.py
new file mode 100644
index 0000000..2ecdfbe
--- /dev/null
+++ b/app/pages/metered_weight_forecast.py
@@ -0,0 +1,716 @@
+import streamlit as st
+import plotly.express as px
+import plotly.graph_objects as go
+import pandas as pd
+import numpy as np
+import joblib
+import os
+from datetime import datetime, timedelta
+from app.services.extruder_service import ExtruderService
+from app.services.main_process_service import MainProcessService
+
+# 灏濊瘯瀵煎叆torch锛屽鏋滃け璐ュ垯绂佺敤娣卞害瀛︿範妯″瀷鏀寔
+try:
+ import torch
+ TORCH_AVAILABLE = True
+except ImportError:
+ TORCH_AVAILABLE = False
+
+
+# 绋虫�佽瘑鍒被
+class SteadyStateDetector:
+ def __init__(self):
+ pass
+
+ def detect_steady_state(self, df, weight_col='绫抽噸', window_size=20, std_threshold=0.5, duration_threshold=60):
+ """
+ 绋虫�佽瘑鍒�昏緫锛氭爣璁扮背閲嶆暟鎹腑鐨勭ǔ鎬佹
+ :param df: 鍖呭惈绫抽噸鏁版嵁鐨勬暟鎹
+ :param weight_col: 绫抽噸鍒楀悕
+ :param window_size: 婊戝姩绐楀彛澶у皬锛堢锛�
+ :param std_threshold: 鏍囧噯宸槇鍊�
+ :param duration_threshold: 绋虫�佹寔缁椂闂撮槇鍊硷紙绉掞級
+ :return: 鍖呭惈绋虫�佹爣璁扮殑鏁版嵁妗嗗拰绋虫�佷俊鎭�
+ """
+ if df is None or df.empty:
+ return df, []
+
+ # 纭繚鏃堕棿鍒楁槸datetime绫诲瀷
+ df['time'] = pd.to_datetime(df['time'])
+
+ # 璁＄畻婊氬姩缁熻閲�
+ df['rolling_std'] = df[weight_col].rolling(window=window_size, min_periods=5).std()
+ df['rolling_mean'] = df[weight_col].rolling(window=window_size, min_periods=5).mean()
+
+ # 璁＄畻娉㈠姩鑼冨洿
+ df['fluctuation_range'] = (df['rolling_std'] / df['rolling_mean']) * 100
+ df['fluctuation_range'] = df['fluctuation_range'].fillna(0)
+
+ # 鏍囪绋虫�佺偣
+ df['is_steady'] = 0
+ steady_condition = (
+ (df['fluctuation_range'] < std_threshold) &
+ (df[weight_col] >= 0.1)
+ )
+ df.loc[steady_condition, 'is_steady'] = 1
+
+ # 璇嗗埆杩炵画绋虫�佹
+ steady_segments = []
+ current_segment = {}
+
+ for i, row in df.iterrows():
+ if row['is_steady'] == 1:
+ if not current_segment:
+ current_segment = {
+ 'start_time': row['time'],
+ 'start_idx': i,
+ 'weights': [row[weight_col]]
+ }
+ else:
+ current_segment['weights'].append(row[weight_col])
+ else:
+ if current_segment:
+ current_segment['end_time'] = df.loc[i-1, 'time'] if i > 0 else df.loc[i, 'time']
+ current_segment['end_idx'] = i-1
+ duration = (current_segment['end_time'] - current_segment['start_time']).total_seconds()
+
+ if duration >= duration_threshold:
+ weights_array = np.array(current_segment['weights'])
+ current_segment['duration'] = duration
+ current_segment['mean_weight'] = np.mean(weights_array)
+ current_segment['std_weight'] = np.std(weights_array)
+ current_segment['min_weight'] = np.min(weights_array)
+ current_segment['max_weight'] = np.max(weights_array)
+ current_segment['fluctuation_range'] = (current_segment['std_weight'] / current_segment['mean_weight']) * 100
+
+ # 璁＄畻缃俊搴�
+ confidence = 100 - (current_segment['fluctuation_range'] / std_threshold) * 50
+ confidence = max(50, min(100, confidence))
+ current_segment['confidence'] = confidence
+
+ steady_segments.append(current_segment)
+
+ current_segment = {}
+
+ # 澶勭悊鏈�鍚庝竴涓ǔ鎬佹
+ if current_segment:
+ current_segment['end_time'] = df['time'].iloc[-1]
+ current_segment['end_idx'] = len(df) - 1
+ duration = (current_segment['end_time'] - current_segment['start_time']).total_seconds()
+
+ if duration >= duration_threshold:
+ weights_array = np.array(current_segment['weights'])
+ current_segment['duration'] = duration
+ current_segment['mean_weight'] = np.mean(weights_array)
+ current_segment['std_weight'] = np.std(weights_array)
+ current_segment['min_weight'] = np.min(weights_array)
+ current_segment['max_weight'] = np.max(weights_array)
+ current_segment['fluctuation_range'] = (current_segment['std_weight'] / current_segment['mean_weight']) * 100
+
+ confidence = 100 - (current_segment['fluctuation_range'] / std_threshold) * 50
+ confidence = max(50, min(100, confidence))
+ current_segment['confidence'] = confidence
+
+ steady_segments.append(current_segment)
+
+ # 鍦ㄦ暟鎹涓爣璁板畬鏁寸殑绋虫�佹
+ for segment in steady_segments:
+ df.loc[segment['start_idx']:segment['end_idx'], 'is_steady'] = 1
+
+ return df, steady_segments
+
+
+def show_metered_weight_forecast():
+ # 鍒濆鍖栨湇鍔�
+ extruder_service = ExtruderService()
+ main_process_service = MainProcessService()
+
+ # 椤甸潰鏍囬
+ st.title("绫抽噸棰勬祴鍒嗘瀽")
+
+ # 鍒濆鍖栦細璇濈姸鎬�
+ if 'forecast_start_date' not in st.session_state:
+ st.session_state['forecast_start_date'] = datetime.now().date() - timedelta(days=7)
+ if 'forecast_end_date' not in st.session_state:
+ st.session_state['forecast_end_date'] = datetime.now().date()
+ if 'forecast_quick_select' not in st.session_state:
+ st.session_state['forecast_quick_select'] = "鏈�杩�7澶�"
+ if 'selected_model' not in st.session_state:
+ st.session_state['selected_model'] = None
+ if 'selected_model_file' not in st.session_state:
+ st.session_state['selected_model_file'] = None
+ if 'forecast_use_steady_only' not in st.session_state:
+ st.session_state['forecast_use_steady_only'] = True
+ if 'forecast_steady_window' not in st.session_state:
+ st.session_state['forecast_steady_window'] = 20
+ if 'forecast_steady_threshold' not in st.session_state:
+ st.session_state['forecast_steady_threshold'] = 1.5
+
+ # 瀹氫箟鍥炶皟鍑芥暟
+ def update_dates(qs):
+ st.session_state['forecast_quick_select'] = qs
+ today = datetime.now().date()
+ if qs == "浠婂ぉ":
+ st.session_state['forecast_start_date'] = today
+ st.session_state['forecast_end_date'] = today
+ elif qs == "鏈�杩�3澶�":
+ st.session_state['forecast_start_date'] = today - timedelta(days=3)
+ st.session_state['forecast_end_date'] = today
+ elif qs == "鏈�杩�7澶�":
+ st.session_state['forecast_start_date'] = today - timedelta(days=7)
+ st.session_state['forecast_end_date'] = today
+ elif qs == "鏈�杩�30澶�":
+ st.session_state['forecast_start_date'] = today - timedelta(days=30)
+ st.session_state['forecast_end_date'] = today
+
+ def on_date_change():
+ st.session_state['forecast_quick_select'] = "鑷畾涔�"
+
+ # 鏌ヨ鏉′欢鍖哄煙
+ with st.expander("馃攳 鏁版嵁閫夋嫨", expanded=True):
+ # 娣诲姞鑷畾涔� CSS 瀹炵幇鍝嶅簲寮忔崲琛�
+ st.markdown("""
+ <style>
+ /* 寮哄埗鍒楀鍣ㄦ崲琛� */
+ [data-testid="stExpander"] [data-testid="column"] {
+ flex: 1 1 120px !important;
+ min-width: 120px !important;
+ }
+ /* 閽堝鏃ユ湡杈撳叆妗嗗垪绋嶅井鍔犲涓�鐐� */
+ @media (min-width: 768px) {
+ [data-testid="stExpander"] [data-testid="column"]:nth-child(6),
+ [data-testid="stExpander"] [data-testid="column"]:nth-child(7) {
+ flex: 2 1 180px !important;
+ min-width: 180px !important;
+ }
+ }
+ </style>
+ """, unsafe_allow_html=True)
+
+ # 鍒涘缓甯冨眬
+ cols = st.columns([1, 1, 1, 1, 1, 1.5, 1.5, 1])
+
+ options = ["浠婂ぉ", "鏈�杩�3澶�", "鏈�杩�7澶�", "鏈�杩�30澶�", "鑷畾涔�"]
+ for i, option in enumerate(options):
+ with cols[i]:
+ # 鏍规嵁褰撳墠閫夋嫨鐘舵�佸喅瀹氭寜閽被鍨�
+ button_type = "primary" if st.session_state['forecast_quick_select'] == option else "secondary"
+ if st.button(option, key=f"btn_forecast_{option}", width='stretch', type=button_type):
+ update_dates(option)
+ st.rerun()
+
+ with cols[5]:
+ start_date = st.date_input(
+ "寮�濮嬫棩鏈�",
+ label_visibility="collapsed",
+ key="forecast_start_date",
+ on_change=on_date_change
+ )
+
+ with cols[6]:
+ end_date = st.date_input(
+ "缁撴潫鏃ユ湡",
+ label_visibility="collapsed",
+ key="forecast_end_date",
+ on_change=on_date_change
+ )
+
+ with cols[7]:
+ query_button = st.button("馃殌 鏌ヨ鏁版嵁", key="forecast_query", width='stretch')
+
+ # 杞崲涓篸atetime瀵硅薄
+ start_dt = datetime.combine(start_date, datetime.min.time())
+ end_dt = datetime.combine(end_date, datetime.max.time())
+
+ # 妯″瀷閫夋嫨鍖哄煙
+ with st.expander("馃搧 妯″瀷閫夋嫨", expanded=True):
+ # 鍒涘缓妯″瀷鐩綍锛堝鏋滀笉瀛樺湪锛�
+ model_dir = "saved_models"
+ os.makedirs(model_dir, exist_ok=True)
+
+ # 鑾峰彇鎵�鏈夊凡淇濆瓨鐨勬ā鍨嬫枃浠�
+ model_files = [f for f in os.listdir(model_dir) if f.endswith('.joblib')]
+ model_files.sort(reverse=True) # 鏈�鏂扮殑妯″瀷鎺掑湪鍓嶉潰
+
+ if not model_files:
+ st.warning("灏氭湭淇濆瓨浠讳綍妯″瀷锛岃鍏堣缁冩ā鍨嬪苟淇濆瓨銆�")
+ else:
+ # 妯″瀷閫夋嫨涓嬫媺妗�
+ selected_model_file = st.selectbox(
+ "閫夋嫨宸蹭繚瀛樼殑妯″瀷",
+ options=model_files,
+ help="閫夋嫨瑕佺敤浜庨娴嬬殑妯″瀷鏂囦欢",
+ key="forecast_selected_model"
+ )
+
+ # 鍔犺浇骞舵樉绀烘ā鍨嬩俊鎭�
+ if selected_model_file:
+ model_path = os.path.join(model_dir, selected_model_file)
+ model_info = joblib.load(model_path)
+
+ # 鏄剧ず妯″瀷鍩烘湰淇℃伅
+ st.subheader("馃搳 妯″瀷淇℃伅")
+ info_cols = st.columns(2)
+
+ with info_cols[0]:
+ st.metric("妯″瀷绫诲瀷", model_info['model_type'])
+ st.metric("鍒涘缓鏃堕棿", model_info['created_at'].strftime('%Y-%m-%d %H:%M:%S'))
+ st.metric("浣跨敤绋虫�佹暟鎹�", "鏄�" if model_info.get('use_steady_data', False) else "鍚�")
+
+ with info_cols[1]:
+ st.metric("R虏 寰楀垎", f"{model_info['r2_score']:.4f}")
+ st.metric("鍧囨柟璇樊 (MSE)", f"{model_info['mse']:.6f}")
+ st.metric("鍧囨柟鏍硅宸� (RMSE)", f"{model_info['rmse']:.6f}")
+
+ # 鏄剧ず妯″瀷鐗瑰緛
+ st.write("馃攽 妯″瀷浣跨敤鐨勭壒寰�:")
+ st.code(", ".join(model_info['features']))
+
+ # 濡傛灉鏄繁搴﹀涔犳ā鍨嬶紝鏄剧ず搴忓垪闀垮害
+ if 'sequence_length' in model_info:
+ st.metric("搴忓垪闀垮害", model_info['sequence_length'])
+
+ # 淇濆瓨妯″瀷淇℃伅鍒颁細璇濈姸鎬�
+ st.session_state['selected_model'] = model_info
+ st.session_state['selected_model_file'] = selected_model_file
+
+ # 绋虫�佽瘑鍒厤缃�
+ st.markdown("---")
+ st.write("鈿栵笍 **绋虫�佽瘑鍒厤缃�**")
+
+ steady_cols = st.columns(3)
+ with steady_cols[0]:
+ st.checkbox(
+ "浠呴娴嬬ǔ鎬佹暟鎹�",
+ value=st.session_state['forecast_use_steady_only'],
+ key="forecast_use_steady_only",
+ help="鍚敤鍚庯紝鍙澶勪簬绋虫�佹椂娈电殑鏁版嵁杩涜绫抽噸棰勬祴"
+ )
+
+ with steady_cols[1]:
+ st.slider(
+ "婊戝姩绐楀彛澶у皬 (绉�)",
+ min_value=5,
+ max_value=60,
+ value=st.session_state['forecast_steady_window'],
+ step=5,
+ key="forecast_steady_window",
+ help="鐢ㄤ簬绋虫�佽瘑鍒殑婊戝姩绐楀彛澶у皬"
+ )
+
+ with steady_cols[2]:
+ st.slider(
+ "娉㈠姩闃堝�� (%)",
+ min_value=0.1,
+ max_value=2.0,
+ value=st.session_state['forecast_steady_threshold'],
+ step=0.1,
+ key="forecast_steady_threshold",
+ help="绋虫�佽瘑鍒殑娉㈠姩鑼冨洿闃堝��"
+ )
+
+ # 棰勬祴鍔熻兘鍖哄煙
+ st.subheader("馃敭 绫抽噸棰勬祴")
+
+ if query_button and st.session_state['selected_model']:
+ with st.spinner("姝ｅ湪鑾峰彇鏁版嵁骞惰繘琛岄娴�..."):
+ # 1. 鑾峰彇瀹屾暣鐨勬尋鍑烘満鏁版嵁
+ df_extruder_full = extruder_service.get_extruder_data(start_dt, end_dt)
+
+ # 2. 鑾峰彇涓绘祦绋嬫帶鍒舵暟鎹�
+ df_main_speed = main_process_service.get_cutting_setting_data(start_dt, end_dt)
+ df_temp = main_process_service.get_temperature_control_data(start_dt, end_dt)
+
+ # 妫�鏌ユ槸鍚︽湁鏁版嵁
+ has_data = any([
+ df_extruder_full is not None and not df_extruder_full.empty,
+ df_main_speed is not None and not df_main_speed.empty,
+ df_temp is not None and not df_temp.empty
+ ])
+
+ if not has_data:
+ st.warning("鎵�閫夋椂闂存鍐呮湭鎵惧埌浠讳綍鏁版嵁锛岃灏濊瘯璋冩暣鏌ヨ鏉′欢銆�")
+ else:
+ # 鏁版嵁鏁村悎涓庨澶勭悊
+ def integrate_data(df_extruder_full, df_main_speed, df_temp):
+ # 纭繚鎸ゅ嚭鏈烘暟鎹瓨鍦�
+ if df_extruder_full is None or df_extruder_full.empty:
+ return None
+
+ # 鍒涘缓鍙寘鍚背閲嶅拰鏃堕棿鐨勪富鏁版嵁闆�
+ df_merged = df_extruder_full[['time', 'metered_weight', 'screw_speed_actual', 'head_pressure']].copy()
+
+
+ # 鏁村悎涓绘祦绋嬫暟鎹�
+ if df_main_speed is not None and not df_main_speed.empty:
+ df_main_speed = df_main_speed[['time', 'process_main_speed']]
+ df_merged = pd.merge_asof(
+ df_merged.sort_values('time'),
+ df_main_speed.sort_values('time'),
+ on='time',
+ direction='nearest',
+ tolerance=pd.Timedelta('1min')
+ )
+
+ # 鏁村悎娓╁害鏁版嵁
+ if df_temp is not None and not df_temp.empty:
+ temp_cols = ['time', 'nakata_extruder_screw_display_temp',
+ 'nakata_extruder_rear_barrel_display_temp',
+ 'nakata_extruder_front_barrel_display_temp',
+ 'nakata_extruder_head_display_temp']
+ df_temp_subset = df_temp[temp_cols].copy()
+ df_merged = pd.merge_asof(
+ df_merged.sort_values('time'),
+ df_temp_subset.sort_values('time'),
+ on='time',
+ direction='nearest',
+ tolerance=pd.Timedelta('1min')
+ )
+
+ # 閲嶅懡鍚嶅垪浠ユ彁楂樺彲璇绘��
+ df_merged.rename(columns={
+ 'screw_speed_actual': '铻烘潌杞��',
+ 'head_pressure': '鏈哄ご鍘嬪姏',
+ 'process_main_speed': '娴佺▼涓婚��',
+ 'nakata_extruder_screw_display_temp': '铻烘潌娓╁害',
+ 'nakata_extruder_rear_barrel_display_temp': '鍚庢満绛掓俯搴�',
+ 'nakata_extruder_front_barrel_display_temp': '鍓嶆満绛掓俯搴�',
+ 'nakata_extruder_head_display_temp': '鏈哄ご娓╁害'
+ }, inplace=True)
+
+ # 娓呯悊鏁版嵁
+ df_merged.dropna(subset=['metered_weight'], inplace=True)
+
+ return df_merged
+
+ # 鎵ц鏁版嵁鏁村悎
+ df_analysis = integrate_data(df_extruder_full, df_main_speed, df_temp)
+
+ if df_analysis is None or df_analysis.empty:
+ st.warning("鏁版嵁鏁村悎澶辫触锛岃妫�鏌ユ暟鎹川閲忔垨璋冩暣鏃堕棿鑼冨洿銆�")
+ else:
+ # 閲嶅懡鍚嶇背閲嶅垪
+ df_analysis.rename(columns={'metered_weight': '绫抽噸'}, inplace=True)
+
+ # 绋虫�佽瘑鍒�
+ steady_detector = SteadyStateDetector()
+
+ # 鑾峰彇绋虫�佽瘑鍒弬鏁�
+ use_steady_only = st.session_state.get('forecast_use_steady_only', True)
+ steady_window = st.session_state.get('forecast_steady_window', 20)
+ steady_threshold = st.session_state.get('forecast_steady_threshold', 0.5)
+
+ # 鎵ц绋虫�佽瘑鍒�
+ df_analysis_with_steady, steady_segments = steady_detector.detect_steady_state(
+ df_analysis,
+ weight_col='绫抽噸',
+ window_size=steady_window,
+ std_threshold=steady_threshold
+ )
+
+ # 鏇存柊df_analysis涓哄寘鍚ǔ鎬佹爣璁扮殑鏁版嵁
+ df_analysis = df_analysis_with_steady
+
+ # 鏄剧ず绋虫�佺粺璁′俊鎭�
+ total_data = len(df_analysis)
+ steady_data = len(df_analysis[df_analysis['is_steady'] == 1])
+ steady_ratio = (steady_data / total_data * 100) if total_data > 0 else 0
+
+ st.subheader("馃搳 绋虫�佹暟鎹粺璁�")
+ stats_cols = st.columns(4)
+ stats_cols[0].metric("鎬绘暟鎹噺", total_data)
+ stats_cols[1].metric("绋虫�佹暟鎹噺", steady_data)
+ stats_cols[2].metric("绋虫�佹暟鎹瘮渚�", f"{steady_ratio:.1f}%")
+ stats_cols[3].metric("绋虫�佹鏁伴噺", len(steady_segments))
+
+ # 鑾峰彇妯″瀷淇℃伅
+ model_info = st.session_state['selected_model']
+ required_features = model_info['features']
+
+ # 妫�鏌ユ墍鏈夊繀闇�鐨勭壒寰佹槸鍚﹀湪鏁版嵁涓�
+ missing_features = [f for f in required_features if f not in df_analysis.columns]
+ if missing_features:
+ st.warning(f"鏁版嵁涓己灏戜互涓嬬壒寰�: {', '.join(missing_features)}")
+ else:
+ # 鍑嗗鎵�鏈夋暟鎹敤浜庢樉绀�
+ df_all = df_analysis.dropna(subset=required_features + ['绫抽噸']).copy()
+
+ if len(df_all) == 0:
+ st.warning("娌℃湁瓒冲鐨勬湁鏁堟暟鎹繘琛岄娴嬶紝璇疯皟鏁存椂闂磋寖鍥存垨妫�鏌ユ暟鎹川閲忋��")
+ else:
+ # 鏍规嵁閰嶇疆鍐冲畾鏄惁鍙娇鐢ㄧǔ鎬佹暟鎹繘琛岄娴�
+ if use_steady_only:
+ df_pred_steady = df_all[df_all['is_steady'] == 1].copy()
+ if len(df_pred_steady) > 0:
+ df_pred = df_pred_steady
+ st.info(f"宸插惎鐢ㄧǔ鎬佽繃婊わ紝浣跨敤 {len(df_pred)} 鏉＄ǔ鎬佹暟鎹繘琛岄娴�")
+ else:
+ df_pred = df_all.copy()
+ st.warning("鏈壘鍒扮ǔ鎬佹暟鎹紝灏嗕娇鐢ㄦ墍鏈夋暟鎹繘琛岄娴�")
+ else:
+ df_pred = df_all.copy()
+
+ # 鎵ц棰勬祴 - 鍙閫夊畾鐨勬暟鎹紙绋虫�佹垨鍏ㄩ儴锛夎繘琛岄娴�
+ X_pred = df_pred[required_features]
+ predicted_weights = []
+
+ # 鑾峰彇妯″瀷
+ model = model_info['model']
+
+ # 妫�鏌ユā鍨嬬被鍨嬪苟鎵ц棰勬祴
+ if model_info['model_type'] in ['LSTM', 'GRU', 'BiLSTM']:
+ # 娣卞害瀛︿範妯″瀷棰勬祴
+ if not TORCH_AVAILABLE:
+ st.error("PyTorch 鏈畨瑁咃紝鏃犳硶浣跨敤娣卞害瀛︿範妯″瀷杩涜棰勬祴銆�")
+ st.stop()
+
+ # 鏁版嵁鏍囧噯鍖�
+ scaler_X = model_info['scaler_X']
+ scaler_y = model_info['scaler_y']
+ X_scaled = scaler_X.transform(X_pred)
+
+ # 鑾峰彇搴忓垪闀垮害
+ sequence_length = model_info['sequence_length']
+
+ # 涓烘繁搴﹀涔犳ā鍨嬪垱寤哄簭鍒�
+ def create_sequences(data, seq_length):
+ sequences = []
+ for i in range(len(data) - seq_length + 1):
+ seq = data[i:i+seq_length]
+ sequences.append(seq)
+ return np.array(sequences)
+
+ X_sequences = create_sequences(X_scaled, sequence_length)
+
+ # 杞崲涓篜yTorch寮犻噺
+ import torch
+ device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+ X_tensor = torch.tensor(X_sequences, dtype=torch.float32).to(device)
+
+ # 棰勬祴
+ model.eval()
+ with torch.no_grad():
+ y_pred_scaled_tensor = model(X_tensor)
+ y_pred_scaled = y_pred_scaled_tensor.cpu().numpy().ravel()
+
+ # 鍙嶅綊涓�鍖�
+ predicted = scaler_y.inverse_transform(y_pred_scaled.reshape(-1, 1)).ravel()
+
+ # 鐢变簬搴忓垪棰勬祴锛屾垜浠渶瑕佸～鍏呭墠闈㈢殑缂哄け鍊�
+ predicted_weights = [np.nan] * (sequence_length - 1) + list(predicted)
+
+ elif model_info['model_type'] in ['SVR', 'MLP']:
+ # 鏀寔鍚戦噺鏈烘垨澶氬眰鎰熺煡鍣ㄩ娴�
+ # 鏁版嵁鏍囧噯鍖�
+ scaler_X = model_info['scaler_X']
+ scaler_y = model_info['scaler_y']
+ X_scaled = scaler_X.transform(X_pred)
+
+ # 棰勬祴
+ y_pred_scaled = model.predict(X_scaled)
+
+ # 鍙嶅綊涓�鍖�
+ predicted_weights = scaler_y.inverse_transform(y_pred_scaled.reshape(-1, 1)).ravel()
+
+ else:
+ # 鍏朵粬妯″瀷锛堝闅忔満妫灄銆佹搴︽彁鍗囥�佺嚎鎬у洖褰掔瓑锛�
+ predicted_weights = model.predict(X_pred)
+
+ # 灏嗛娴嬬粨鏋滄坊鍔犲埌鏁版嵁妗嗕腑
+ df_pred['棰勬祴绫抽噸'] = predicted_weights
+
+ # 纭繚鏃堕棿鍒楁槸datetime绫诲瀷
+ df_pred['time'] = pd.to_datetime(df_pred['time'])
+
+ # 鏁版嵁瀵规瘮鍔熻兘
+ st.subheader("馃搳 棰勬祴缁撴灉瀵规瘮鍒嗘瀽")
+
+ # 璁＄畻棰勬祴璇樊
+ df_pred['璇樊'] = df_pred['棰勬祴绫抽噸'] - df_pred['绫抽噸']
+ df_pred['缁濆璇樊'] = abs(df_pred['璇樊'])
+ df_pred['鐩稿璇樊'] = (df_pred['缁濆璇樊'] / df_pred['绫抽噸']) * 100
+
+ # 鏄剧ず璇樊缁熻淇℃伅
+ error_stats = df_pred.dropna(subset=['棰勬祴绫抽噸']).describe()
+
+ stats_cols = st.columns(3)
+ with stats_cols[0]:
+ st.metric("骞冲潎瀹為檯绫抽噸", f"{error_stats['绫抽噸']['mean']:.4f} Kg/m")
+ st.metric("骞冲潎棰勬祴绫抽噸", f"{error_stats['棰勬祴绫抽噸']['mean']:.4f} Kg/m")
+ with stats_cols[1]:
+ st.metric("骞冲潎缁濆璇樊", f"{error_stats['缁濆璇樊']['mean']:.4f} Kg/m")
+ st.metric("鏈�澶х粷瀵硅宸�", f"{error_stats['缁濆璇樊']['max']:.4f} Kg/m")
+ with stats_cols[2]:
+ st.metric("骞冲潎鐩稿璇樊", f"{error_stats['鐩稿璇樊']['mean']:.2f}%")
+ st.metric("鏈�澶х浉瀵硅宸�", f"{error_stats['鐩稿璇樊']['max']:.2f}%")
+
+ # 鍙鍖栧睍绀�
+ st.subheader("馃搱 绫抽噸瓒嬪娍瀵规瘮")
+
+ # 鍒涘缓瓒嬪娍鍥� - 浣跨敤鎵�鏈夋暟鎹甦f_all杩涜鏄剧ず
+ fig = go.Figure()
+
+ # 纭繚鏃堕棿鍒楁槸datetime绫诲瀷
+ df_all['time'] = pd.to_datetime(df_all['time'])
+
+ # # 娣诲姞瀹炴椂绫抽噸鏁版嵁鐐癸紙绋虫�佹暟鎹敤钃濊壊锛岄潪绋虫�佹暟鎹敤鐏拌壊锛�
+ # if 'is_steady' in df_all.columns:
+ # # 绋虫�佹暟鎹� - 浣跨敤鐐规樉绀�
+ # steady_data = df_all[df_all['is_steady'] == 1]
+ # non_steady_data = df_all[df_all['is_steady'] == 0]
+
+ # if len(steady_data) > 0:
+ # fig.add_trace(go.Scatter(
+ # x=steady_data['time'],
+ # y=steady_data['绫抽噸'],
+ # name='瀹炴椂绫抽噸锛堢ǔ鎬侊級',
+ # mode='markers',
+ # marker=dict(color='blue', size=3),
+ # hovertemplate='鏃堕棿: %{x}<br>瀹炴椂绫抽噸锛堢ǔ鎬侊級: %{y:.4f} Kg/m<extra></extra>'
+ # ))
+
+ # # 闈炵ǔ鎬佹暟鎹篃鏄剧ず锛屼絾涓嶈繘琛岄娴�
+ # if len(non_steady_data) > 0:
+ # fig.add_trace(go.Scatter(
+ # x=non_steady_data['time'],
+ # y=non_steady_data['绫抽噸'],
+ # name='瀹炴椂绫抽噸锛堥潪绋虫�侊級',
+ # mode='markers',
+ # marker=dict(color='lightgray', size=3),
+ # hovertemplate='鏃堕棿: %{x}<br>瀹炴椂绫抽噸锛堥潪绋虫�侊級: %{y:.4f} Kg/m<extra></extra>'
+ # ))
+ # else:
+ # 濡傛灉娌℃湁绋虫�佹爣璁帮紝鏄剧ず鎵�鏈夋暟鎹偣
+ fig.add_trace(go.Scatter(
+ x=df_all['time'],
+ y=df_all['绫抽噸'],
+ name='瀹炴椂绫抽噸',
+ mode='lines',
+ line=dict(color='blue', width=1.5),
+ # hovertemplate='鏃堕棿: %{x}<br>瀹炴椂绫抽噸: %{y:.4f} Kg/m<extra></extra>'
+ ))
+
+ # 娣诲姞棰勬祴绫抽噸鏇茬嚎 - 鍙棰勬祴鐨勬暟鎹紙绋虫�佹垨鍏ㄩ儴锛夋樉绀�
+ fig.add_trace(go.Scatter(
+ x=df_pred['time'],
+ y=df_pred['棰勬祴绫抽噸'],
+ name='棰勬祴绫抽噸',
+ mode='lines',
+ line=dict(color='red', width=2, dash='dash'),
+ marker=dict(size=3),
+ # hovertemplate='鏃堕棿: %{x}<br>棰勬祴绫抽噸: %{y:.4f} Kg/m<extra></extra>'
+ ))
+
+ # 娣诲姞鎵�鏈夋尋鍑烘満鍙傛暟鏇茬嚎 - 浣跨敤鎵�鏈夋暟鎹�
+ colors = ['green', 'orange', 'purple', 'brown', 'pink', 'gray', 'olive', 'cyan', 'magenta', 'yellow', 'lime', 'teal']
+ for i, feature in enumerate(required_features):
+ # 涓烘瘡涓壒寰佸垎閰嶄笉鍚岀殑棰滆壊
+ color = colors[i % len(colors)]
+
+ # 纭繚鐗瑰緛瀛樺湪浜庢墍鏈夋暟鎹腑
+ if feature in df_all.columns:
+ fig.add_trace(go.Scatter(
+ x=df_all['time'],
+ y=df_all[feature],
+ name=feature,
+ mode='lines',
+ line=dict(color=color, width=1.5),
+ yaxis=f'y{i+2}',
+ # hovertemplate=f'鏃堕棿: %{{x}}<br>{feature}: %{{y}}<extra></extra>'
+ ))
+
+ # 閰嶇疆鍥捐〃甯冨眬
+ layout = {
+ 'title': '绫抽噸棰勬祴涓庡疄鏃舵暟鎹姣�',
+ 'xaxis': {
+ 'title': '鏃堕棿',
+ 'rangeslider': {'visible': True},
+ 'type': 'date',
+ 'tickformat': '%Y-%m-%d %H:%M'
+ },
+ 'yaxis': {
+ 'title': '绫抽噸 (Kg/m)',
+ 'title_font': {'color': 'blue'},
+ 'tickfont': {'color': 'blue'},
+ 'side': 'left',
+ 'fixedrange': False # 鍏佽y杞寸缉鏀�
+ },
+ 'legend': {
+ 'orientation': 'h',
+ 'yanchor': 'bottom',
+ 'y': 1.02,
+ 'xanchor': 'right',
+ 'x': 1
+ },
+ 'height': 600,
+ 'margin': {'l': 100, 'r': 200, 't': 100, 'b': 100},
+ 'hovermode': 'x unified'
+ }
+
+ # 娣诲姞棰濆鐨剏杞撮厤缃� - 涓烘墍鏈夌壒寰佸垱寤簓杞�
+ for i, feature in enumerate(required_features):
+ layout[f'yaxis{i+2}'] = {
+ 'title': feature,
+ 'title_font': {'color': colors[i % len(colors)]},
+ 'tickfont': {'color': colors[i % len(colors)]},
+ 'overlaying': 'y',
+ 'side': 'right',
+ 'anchor': 'free',
+ 'position': 1 - (i+1)*0.08,
+ 'fixedrange': False # 鍏佽y杞寸缉鏀�
+ }
+
+ fig.update_layout(layout)
+
+ # 鏄剧ず瓒嬪娍鍥� - 鍚敤瀹屾暣鐨勪氦浜掑姛鑳�
+ st.plotly_chart(fig, use_container_width=True, config={
+ 'scrollZoom': True,
+ 'displayModeBar': True,
+ 'modeBarButtonsToAdd': ['pan2d', 'select2d', 'lasso2d', 'resetScale2d'],
+ 'displaylogo': False
+ })
+
+ # 璇樊鍒嗘瀽鍥�
+ st.subheader("馃搲 棰勬祴璇樊鍒嗘瀽")
+
+ # 鍒涘缓璇樊鍒嗗竷鐩存柟鍥�
+ fig_error = px.histogram(df_pred.dropna(subset=['鐩稿璇樊']), x='鐩稿璇樊', nbins=50,
+ title='棰勬祴鐩稿璇樊鍒嗗竷',
+ labels={'鐩稿璇樊': '鐩稿璇樊 (%)'})
+ fig_error.update_layout(
+ xaxis_title='鐩稿璇樊 (%)',
+ yaxis_title='棰戞',
+ height=400
+ )
+ st.plotly_chart(fig_error, use_container_width=True)
+
+ # 鏁版嵁棰勮
+ st.subheader("馃攳 鏁版嵁棰勮")
+ preview_columns = ['time', '绫抽噸', '棰勬祴绫抽噸', '璇樊', '缁濆璇樊', '鐩稿璇樊']
+ if 'is_steady' in df_pred.columns:
+ preview_columns.append('is_steady')
+ preview_columns.extend(required_features)
+ st.dataframe(df_pred[preview_columns].head(20),
+ use_container_width=True)
+
+ # 瀵煎嚭鏁版嵁
+ st.subheader("馃捑 瀵煎嚭鏁版嵁")
+ # 灏嗘暟鎹浆鎹负CSV鏍煎紡
+ csv = df_pred.to_csv(index=False)
+ # 鍒涘缓涓嬭浇鎸夐挳
+ st.download_button(
+ label="瀵煎嚭棰勬祴缁撴灉鏁版嵁 (CSV)",
+ data=csv,
+ file_name=f"metered_weight_forecast_{datetime.now().strftime('%Y%m%d_%H%M%S')}.csv",
+ mime="text/csv",
+ help="鐐瑰嚮鎸夐挳瀵煎嚭棰勬祴缁撴灉鏁版嵁"
+ )
+ elif query_button:
+ st.warning("璇峰厛閫夋嫨涓�涓ā鍨嬨��")
+ else:
+ st.info("璇烽�夋嫨鏃堕棿鑼冨洿鍜屾ā鍨嬶紝鐒跺悗鐐瑰嚮'鏌ヨ鏁版嵁'鎸夐挳寮�濮嬮娴嬪垎鏋愩��")
+
+
+# 椤甸潰鍏ュ彛
+if __name__ == "__main__":
+ show_metered_weight_forecast()
\ No newline at end of file
--
Gitblit v1.9.3