update dashbord with fault color

dashboard.py

@@ -11,6 +11,7 @@ import mqtt_client
 import time
 from src.rtu.RTUPipeline import RTUPipeline
 from src.rtu.RTUAnomalizer1 import RTUAnomalizer1
+from src.rtu.RTUAnomalizer2 import RTUAnomalizer2
 import plotly.express as px
 
 rtu_data_pipeline = RTUPipeline(
@@ -38,15 +39,15 @@ rtu_anomalizers.append(
 )
 
 rtu_anomalizers.append(
-    RTUAnomalizer1(
-        prediction_model_path="src/rtu/models/lstm_2rtu_smooth_04.keras",
+    RTUAnomalizer2(
+        prediction_model_path="src/rtu/models/lstm_2rtu_smooth_03.keras",
         clustering_model_paths=[
-            "src/rtu/models/kmeans_rtu_1.pkl",
-            "src/rtu/models/kmeans_rtu_2.pkl",
+            "src/rtu/models/kmeans_rtu_3.pkl",
+            "src/rtu/models/kmeans_rtu_4.pkl",
         ],
         pca_model_paths=[
-            "src/rtu/models/pca_rtu_1.pkl",
-            "src/rtu/models/pca_rtu_2.pkl",
+            "src/rtu/models/pca_rtu_3.pkl",
+            "src/rtu/models/pca_rtu_4.pkl",
         ],
         num_inputs=rtu_data_pipeline.num_inputs,
         num_outputs=rtu_data_pipeline.num_outputs,
@@ -159,12 +160,21 @@ for i in range(4):
 
 
 # Temperatures streaming and updates
-def update_status_boxes(df):
+def update_status_boxes(df,fault):
     for i in range(4):
         sa_temp = df[f"rtu_00{i+1}_sa_temp"].iloc[-1]
         ra_temp = df[f"rtu_00{i+1}_ra_temp"].iloc[-1]
         rtu_placeholders[i]["sa_temp"].markdown(f"**SA temp:**  {sa_temp} °F")
         rtu_placeholders[i]["ra_temp"].markdown(f"**RA temp:**  {ra_temp} °F")
+        if fault[i]== 1:
+            rtu_placeholders[i]["box"].markdown(
+                f"""
+            <div style='background-color:#ff4d4d;padding:3px;border-radius:5px;margin-bottom:10px'>
+                <h4 style='color:black;text-align:center;'>RTU{i+1}</h4>
+            </div>
+            """,
+                unsafe_allow_html=True,
+            )
 
 
 # Zones
@@ -344,7 +354,7 @@ with st.container():
 distances = []
 
 
-def create_residual_plot(resid_pca_list, rtu_id, lim=8):
+def create_residual_plot(resid_pca_list, distance, rtu_id,lim=8):
     if rtu_id % 2 == 1:
         ax1 = 0
         ax2 = 1
@@ -354,6 +364,7 @@ def create_residual_plot(resid_pca_list, rtu_id, lim=8):
     fig = px.scatter(
         x=resid_pca_list[:, ax1],
         y=resid_pca_list[:, ax2],
+        color=distance,
         labels={"x": "Time", "y": "Residual"},
         width=500,
         height=500,
@@ -372,7 +383,7 @@ def create_residual_plot(resid_pca_list, rtu_id, lim=8):
         hoverlabel_font_color="black",
         hoverlabel_bordercolor="lightgray",
     )
-    fig.update_traces(marker=dict(size=5, color="blue"))
+    # fig.update_traces(marker=dict(size=5, color="blue"))
 
     return fig
 
@@ -401,12 +412,20 @@ while True:
             )
 
         # Loop to update
-        update_status_boxes(df)
+        
 
         dist = None
         resid_pca_list_rtu = None
         resid_pca_list_rtu_2 = None
         resid_pca_list_vav_1 = None
+        rtu_1_distance = None
+        rtu_2_distance = None
+        fault_1 = None
+        fault_2 = None
+        rtu_3_distance = None
+        rtu_4_distance = None
+        fault_3 = None
+        fault_4 = None
 
         df_new1, df_trans1, df_new2, df_trans2 = rtu_data_pipeline.fit(
             pd.DataFrame(mqtt_client.data_list)
@@ -430,7 +449,10 @@ while True:
                 resid_list,
                 resid_pca_list_rtu,
                 dist,
-                over_threshold,
+                rtu_1_distance,
+                rtu_2_distance,
+                fault_1,
+                fault_2
             ) = rtu_anomalizers[0].pipeline(
                 df_new1, df_trans1, rtu_data_pipeline.scaler1
             )
@@ -440,7 +462,10 @@ while True:
                 resid_list_2,
                 resid_pca_list_rtu_2,
                 dist_2,
-                over_threshold_2,
+                rtu_3_distance,
+                rtu_4_distance,
+                fault_3,
+                fault_4
             ) = rtu_anomalizers[1].pipeline(
                 df_new1, df_trans1, rtu_data_pipeline.scaler1
             )
@@ -464,31 +489,31 @@ while True:
                 resid_rtu1_placeholder, resid_rtu2_placeholder = st.columns(2)
                 with resid_rtu1_placeholder:
                     st.subheader("RTU 1 Residuals")
-                    fig = create_residual_plot(resid_pca_list_rtu, rtu_id=1)
+                    fig = create_residual_plot(resid_pca_list_rtu, rtu_1_distance, rtu_id=1)
                     st.plotly_chart(fig)
 
                 with resid_rtu2_placeholder:
                     st.subheader("RTU 2 Residuals")
-                    fig = create_residual_plot(resid_pca_list_rtu, rtu_id=2)
+                    fig = create_residual_plot(resid_pca_list_rtu, rtu_2_distance, rtu_id=2)
                     st.plotly_chart(fig)
 
                 resid_rtu3_placeholder, resid_rtu4_placeholder = st.columns(2)
                 with resid_rtu3_placeholder:
                     st.subheader("RTU 3 Residuals")
-                    fig = create_residual_plot(resid_pca_list_rtu, rtu_id=3)
+                    fig = create_residual_plot(resid_pca_list_rtu, rtu_3_distance, rtu_id=3)
                     st.plotly_chart(fig)
 
                 with resid_rtu4_placeholder:
                     st.subheader("RTU 4 Residuals")
-                    fig = create_residual_plot(resid_pca_list_rtu, rtu_id=4)
+                    fig = create_residual_plot(resid_pca_list_rtu, rtu_4_distance, rtu_id=4)
                     st.plotly_chart(fig)
 
         if resid_pca_list_vav_1 is not None:
-            print(resid_pca_list_vav_1)
+            # print(resid_pca_list_vav_1)
             with resid_vav_placeholder.container():
                 st.subheader("VAV 1 Residuals")
                 fig = create_residual_plot(
-                    np.array(resid_pca_list_vav_1), rtu_id=1, lim=15
+                    np.array(resid_pca_list_vav_1),rtu_4_distance, rtu_id=1, lim=15
                 )
                 st.plotly_chart(fig)
 
@@ -507,5 +532,5 @@ while True:
         #     ax.set_xlabel("Time")
         #     ax.set_ylabel("Energy (kWh)")
         #     st.pyplot(fig)
-
+        update_status_boxes(df,[fault_1,fault_2,fault_3,fault_4])
         mqtt_client.data_list.clear()

src/rtu/RTUAnomalizer1.py

@@ -44,8 +44,8 @@ class RTUAnomalizer1:
             self.initialize_lists()
         )
         
-        self.fault_1 = None
-        self.fault_2 = None
+        self.fault_1 = 0
+        self.fault_2 = 0
 
     def initialize_lists(self, size=30):
         """
@@ -59,7 +59,7 @@ class RTUAnomalizer1:
         """
         initial_values = [[0] * self.num_outputs] * size
         initial_values1 = [[0] * 4] * size
-        initial_values2 = [[0] * 2] * size*2
+        initial_values2 = [[0] * 2] * 60
         return (
             initial_values.copy(),
             initial_values.copy(),
@@ -197,8 +197,8 @@ class RTUAnomalizer1:
                     axis=1,
                 )
             )
-            self.distance_list.pop(0)
-            self.distance_list.append(dist)
+        self.distance_list.pop(0)
+        self.distance_list.append(np.concatenate(dist).tolist())
         resid_pcas = np.array(resid_pcas).flatten().tolist()
         self.resid_pca_list.pop(0)
         self.resid_pca_list.append(resid_pcas)
@@ -206,12 +206,14 @@ class RTUAnomalizer1:
         return np.array(dist)
     def fault_windowing(self):
         rtu_1_dist = np.array(self.distance_list).T[0]>1 #rtu_1_threshold
-        if sum(rtu_1_dist)>80*60: # 80% of the 60 min window
+        rtu_1_dist = [int(x) for x in rtu_1_dist]
+        if sum(rtu_1_dist)>0.8*60: # 80% of the 60 min window
             self.fault_1 = 1
         else:
             self.fault_1 = 0
-        rtu_2_dist = np.array(self.distance_list).T[1]>2.5 #rtu_2_threshold
-        if sum(rtu_2_dist)>80*60: # 80% of the 60 min window
+        rtu_2_dist = np.array(self.distance_list).T[1]>0.5 #rtu_2_threshold
+        rtu_2_dist = [int(x) for x in rtu_2_dist]
+        if sum(rtu_2_dist)>0.05*60: # 80% of the 60 min window
             self.fault_2 = 1
         else:
             self.fault_2 = 0
@@ -235,7 +237,7 @@ class RTUAnomalizer1:
         actual, pred = self.inverse_transform(scaler, pred, df_trans)
         actual_list, pred_list, resid_list = self.update_lists(actual, pred, resid)
         dist = self.calculate_distances(resid)
-        
+        self.fault_windowing()
         return (
             actual_list,
             pred_list,
@@ -243,7 +245,7 @@ class RTUAnomalizer1:
             self.resid_pca_list,
             dist,
             np.array(self.distance_list[30:]).T[0]>1, #rtu_1_threshold
-            np.array(self.distance_list[30:]).T[1]>2.5, #rtu_2_threshold
+            np.array(self.distance_list[30:]).T[1]>0.5, #rtu_2_threshold
             self.fault_1,
             self.fault_2
         )

src/rtu/RTUAnomalizer2.py

@@ -3,7 +3,7 @@ from tensorflow.keras.models import load_model
 import joblib
 
 
-class RTUAnomalizer1:
+class RTUAnomalizer2:
     """
     Class for performing anomaly detection on RTU (Roof Top Unit) data.
     """
@@ -44,8 +44,8 @@ class RTUAnomalizer1:
             self.initialize_lists()
         )
         
-        self.fault_1 = None
-        self.fault_2 = None
+        self.fault_1 = 0
+        self.fault_2 = 0
 
     def initialize_lists(self, size=30):
         """
@@ -197,8 +197,8 @@ class RTUAnomalizer1:
                     axis=1,
                 )
             )
-            self.distance_list.pop(0)
-            self.distance_list.append(dist)
+        self.distance_list.pop(0)
+        self.distance_list.append(np.concatenate(dist).tolist())
         resid_pcas = np.array(resid_pcas).flatten().tolist()
         self.resid_pca_list.pop(0)
         self.resid_pca_list.append(resid_pcas)
@@ -206,12 +206,14 @@ class RTUAnomalizer1:
         return np.array(dist)
     def fault_windowing(self):
         rtu_1_dist = np.array(self.distance_list).T[0]>1.5 #rtu_3_threshold
-        if sum(rtu_1_dist)>80*60: # 80% of the 60 min window
+        rtu_1_dist = [int(x) for x in rtu_1_dist]
+        if sum(rtu_1_dist)>0.8*60: # 80% of the 60 min window
             self.fault_1 = 1
         else:
             self.fault_1 = 0
         rtu_2_dist = np.array(self.distance_list).T[1]>1.5 #rtu_4_threshold
-        if sum(rtu_2_dist)>80*60: # 80% of the 60 min window
+        rtu_2_dist = [int(x) for x in rtu_2_dist]
+        if sum(rtu_2_dist)>0.8*60: # 80% of the 60 min window
             self.fault_2 = 1
         else:
             self.fault_2 = 0
@@ -235,7 +237,7 @@ class RTUAnomalizer1:
         actual, pred = self.inverse_transform(scaler, pred, df_trans)
         actual_list, pred_list, resid_list = self.update_lists(actual, pred, resid)
         dist = self.calculate_distances(resid)
-        
+        self.fault_windowing()
         return (
             actual_list,
             pred_list,