update env logic

models/env.py

@@ -22,6 +22,9 @@ class WgzGym(gym.Env):
         self.episode_length = 24
         self.penalty_coefficient = 50  # 约束惩罚系数
         self.sell_coefficient = 0.5  # 售出利润系数
+        self.a = 0.5
+        self.b = 0.3
+        self.c = 0.2
         self.EC_parameters = kwargs.get('EC_parameters', EC_parameters)  # 电解水制氢器
         self.HST_parameters = kwargs.get('dg_parameters', dg_parameters)  # 储氢罐
 
@@ -69,13 +72,13 @@ class WgzGym(gym.Env):
         current_obs = self._build_state()
         self.EC.step(action[0])  # 执行状态转换，电池当前容量也改变
         self.HST.step(action[1])
-        self.dg2.step(action[2])
+        self.grid.step(action[2])
         price = current_obs[1]
         temper = current_obs[2]
         solar = current_obs[3]
         load = current_obs[4]
         heat = current_obs[5]
-        gym_to_grid = solar + self.EC.get_hydrogen() - load
+        gym_to_grid = solar + self.HST.get_power() - self.EC.current_power - load
 
         # reward = 0.0
         sell_benefit, buy_cost = 0, 0
@@ -90,8 +93,11 @@ class WgzGym(gym.Env):
         hst_cost = self.HST.get_cost()
         ec_cost = self.EC.get_cost(price)
         solar_cost = solar  # 待补充
-        reward = - (hst_cost + ec_cost + solar_cost
-                    + excess_penalty + deficient_penalty - sell_benefit + buy_cost)
+
+        economic_cost = hst_cost + ec_cost + solar_cost - sell_benefit + buy_cost
+        demand_cost = excess_penalty + deficient_penalty
+        eco_benifit = 0
+        reward = - self.a * economic_cost - self.b * demand_cost + self.c * eco_benifit
 
         self.unbalance = gym_to_grid
         final_step_outputs = [self.HST.current_soc, self.EC.current_power, self.grid.current_power]

models/module.py

@@ -25,6 +25,9 @@ class EC:
     def get_carbon(self):
         return self.current_power * self.carbon_reduce
 
+    def get_power_max(self):
+        return self.power_max
+
     def reset(self):
         self.current_power = 0
 
@@ -32,14 +35,16 @@ class EC:
 class HST:
     def __init__(self, params):
         self.current_soc = None
-        self.hydrogen_change = None
+        self.hydrogen_charge = None
         self.capacity = params['capacity']
         self.min_soc = params['min_soc']
         self.max_soc = params['max_soc']
         self.ramp = params['ramp']
         self.lifetime = params['lifetime']
         self.equipment_cost = params['equipment_cost']
-        self.efficiency = params['efficiency']
+        self.charge_efficiency = params['charge_efficiency']
+        self.generate_efficiency = params['generate_efficiency']
+        self.lower_heating_value = params['lower_heating_value']
 
     '''
     储氢罐的充气速率 = 电解水制氢速率 （电解水制氢放的热会满足热水需求?）
@@ -49,11 +54,17 @@ class HST:
     def step(self, action_hst):
         energy = action_hst * self.ramp
         updated_soc = max(self.min_soc, min(self.max_soc, (self.current_soc * self.capacity + energy) / self.capacity))
-        self.hydrogen_change = (updated_capacity - self.current_soc) * self.capacity
+        self.hydrogen_charge = (updated_soc - self.current_soc) * self.capacity
         self.current_soc = updated_soc
 
+    def get_power(self):
+        if self.hydrogen_charge > 0:
+            return self.hydrogen_charge * self.lower_heating_value * self.charge_efficiency * self.generate_efficiency
+        else:
+            return 0
+
     def get_cost(self):
-        cost = self.equipment_cost / self.lifetime * abs(self.hydrogen_change)
+        cost = self.equipment_cost / self.lifetime * abs(self.hydrogen_charge)
         return cost
 
     def reset(self):
@@ -63,10 +74,13 @@ class HST:
 class Grid:
     def __init__(self):
         self.delta = 1
-        self.exchange_ability = 100
+        self.trade_energy = None
 
-    def get_cost(self, current_price, energy_exchange):
-        return current_price * energy_exchange * self.delta
+    def step(self, action_grid):
+        self.trade_energy = (action_grid + 1) / 2 * EC.get_power_max()  # 反标准化
+
+    def get_cost(self, price):
+        return price * self.trade_energy * self.delta
 
     def retrieve_past_price(self):
         result = []

models/parameters.py

@@ -15,5 +15,7 @@ HST_parameters = {
     'max_soc': 0.9,
     'lifetime': 6000,  # hour
     'equipment_cost': 10000,  # yuan
-    'efficiency': 0.95,
+    'charge_efficiency': 0.95,
+    'generate_efficiency': 0.6,
+    'lower_heating_value': 33.33,
 }