Gas unit scheduling method based on multi-stage dual dynamic programming

Abstract

The invention provides a gas unit scheduling method based on multi-stage dual dynamic programming, and belongs to the technical field of electric power scheduling, and the scheduling method specifically comprises the steps: predicting the load condition and new energy power generation of a corresponding power supply region, taking the cost minimization as an optimization target, and constructing a multi-stage scheduling model of the current day of the power supply region; solving the multi-stage scheduling model based on dual dynamic programming to obtain a scheduling strategy of the gas unit; a feedback node is set, when the feedback node is reached, the operation information of the gas unit and the fluctuation information of the prediction result in the corresponding time period are obtained, and the model parameters and the cycle length of the multi-stage scheduling model are correspondingly adjusted; and the adjusted multi-stage adjustment model is solved again, and the scheduling strategy is updated until gas unit scheduling on the current day is completed. According to the method, a feedback regulation mechanism is set while scheduling planning is carried out by using the multi-stage dual dynamic planning algorithm introduced with pre-expansion, and the solving speed and accuracy of gas unit scheduling planning are considered.

Description

Technical Field

[0001] The present invention relates to the technical field of power dispatching, and in particular to a gas unit dispatching method based on multi-stage dual dynamic programming. Background Art

[0002] With the continuous adjustment of energy structure, the mode of coordinated power supply of gas-fired units and new energy power generation is becoming more and more common. Among them, gas-fired units are the basic power supply of the power system. Accurate and efficient power dispatching of gas-fired units is crucial to ensure power supply stability, reduce operating costs and promote the rational use of energy.

[0003] With the expansion of the scale of power systems and the diversification of energy structures, the scenarios of coordinated power supply between gas-fired units and new energy sources are becoming increasingly complex. The amount of calculation required by traditional solution methods has increased dramatically, and the solution efficiency is difficult to meet the needs. When determining the optimal scheduling strategy for large-scale systems, it is necessary to traverse massive combinations, which takes too long and cannot output decision plans in a timely manner.

[0004] In order to solve the above problems, the relevant technology realizes the scheduling planning of gas units by introducing a pre-expanded multi-stage dual dynamic programming algorithm. This algorithm can greatly reduce the computational complexity and significantly improve the solution efficiency by decomposing sub-problems and applying the duality principle, and quickly achieve the global optimality of gas unit scheduling planning.

[0005] However, the model construction of the pre-expanded multi-stage dual dynamic programming is based on simplified assumptions, which deviates from the actual operation of the power system, has poor adaptability to extreme or complex change scenarios, and is highly dependent on data. Data errors, omissions or delays will seriously affect the quality of dual variable calculations and scheduling decisions. For the scheduling of gas-fired units in power systems with large-scale new energy sources, the accuracy of scheduling planning is low. Summary of the invention

[0006] The purpose of the present invention is to overcome the shortcomings of the prior art that when using the pre-expanded multi-stage dual dynamic programming to perform rapid scheduling planning of gas units, the method has poor adaptability to extreme or complex change scenarios, high dependence on data, and low scheduling planning accuracy. A gas unit scheduling method based on multi-stage dual dynamic programming is provided. On the basis of using the pre-expanded multi-stage dual dynamic programming to perform rapid scheduling planning of gas units, the method dynamically adjusts model parameters and cycle lengths by setting feedback nodes and utilizing gas unit operation information and predicted fluctuation information to re-solve and update the scheduling strategy, which can timely adjust the unit operation to ensure power supply, enhance adaptability to extreme or complex change scenarios, further reduce the dependence on data, reduce the impact of data errors through data verification and correction and model adaptive optimization, and improve the accuracy of scheduling planning.

[0007] The purpose of the present invention is achieved through the following technical solutions: The gas unit scheduling method based on multi-stage dual dynamic programming includes: Predict the load conditions and renewable energy power generation in the power supply area corresponding to the gas-fired units; Taking cost minimization as the optimization goal and combining the forecast results, a multi-stage scheduling model for day-ahead unit combination planning and intra-day unit output scheduling in the power supply area is constructed; Solve the multi-stage scheduling model based on dual dynamic programming to obtain the scheduling strategy of the gas unit; Feedback nodes are set according to the day-ahead planning period and intraday scheduling period in the multi-stage scheduling model; When reaching the feedback node, obtain the gas unit operation information and the fluctuation information of the prediction results in the corresponding time period, and adjust the model parameters and cycle length of the multi-stage scheduling model accordingly; The adjusted multi-stage adjustment model is solved again and the dispatching strategy of the gas unit is updated until the dispatching of the gas unit for the day is completed.

[0008] On the basis of using the pre-expanded multi-stage dual dynamic programming to carry out rapid scheduling planning of gas units, by setting feedback nodes, the operation information of the gas units and the predicted fluctuation information are used to verify and correct the load and renewable energy power generation forecast data, so as to reduce the impact of data errors. This method of dynamically adjusting the parameters and cycle length of the multi-stage scheduling model according to real-time information and re-solving and updating the scheduling strategy can timely adjust the unit operation to ensure power supply, enhance the adaptability to extreme or complex change scenarios, further reduce the dependence on data, reduce the impact of data errors and improve the accuracy of scheduling planning through data verification and correction and model adaptive optimization.

[0009] Furthermore, the setting of feedback nodes according to the day-ahead planning period and the intraday scheduling period in the multi-stage scheduling model includes: Determine the current day-ahead planning information and intra-day scheduling information according to the scheduling strategy; Taking the start time of each cycle as the candidate feedback node, the adjustment operation information of the gas unit at each candidate feedback node is calculated according to the corresponding day-ahead planning information and intraday scheduling information; Based on the adjustment operation information, the adjustment cost and adjustment duration are calculated at the alternative feedback nodes corresponding to the day-ahead planning and the alternative feedback nodes corresponding to the intraday scheduling respectively; Calculate the time overlap between the alternative feedback nodes corresponding to the day-ahead planning and intraday scheduling, and set the feedback nodes based on the adjustment cost and adjustment duration.

[0010] Furthermore, the feedback node is set according to the day-ahead planning period and the intraday scheduling period in the multi-stage scheduling model, and further includes: Each time the scheduling strategy is updated, the remaining feedback nodes are reset according to the day-ahead planning period and intraday scheduling period in the adjusted multi-stage scheduling model.

[0011] The feedback nodes are set by comprehensively considering multiple factors such as day-ahead planning information, intraday dispatch information, adjustment duration, time overlap, etc., and the feedback nodes are dynamically adjusted as the dispatch strategy is updated, providing data support that is more in line with real-time conditions for the dispatch decision of the gas unit, ensuring stable and reliable power supply. The dynamic update of feedback nodes can more timely obtain the real-time changes in the operating status of the gas unit, power load and renewable energy power generation. In the face of complex and changeable operating environments and emergencies such as extreme weather and large load fluctuations, it can quickly respond and adjust the dispatch strategy to effectively deal with various uncertainties. It can also capture the key time points of strategy adjustment in a timely manner, reduce unnecessary waiting and invalid operations, make the adjustment operation of the gas unit more efficient, and improve the overall operating efficiency of the power system.

[0012] Furthermore, when the feedback node is reached, the operation information of the gas unit and the fluctuation information of the prediction result in the corresponding time period are obtained, and the model parameters and cycle length of the multi-stage scheduling model are adjusted accordingly, including: Obtain the gas unit operation information, actual load conditions, and actual power generation information of new energy sources in the time period from the previous feedback node to the current feedback node under the current scheduling strategy, and determine the fluctuation information of the prediction results; Identify adjustment requirements based on gas unit operation information and fluctuation information of forecast results; According to the identification results of the adjustment demand, the model parameters and cycle length of the multi-stage scheduling model are adjusted in combination with the operation information of the gas units and the fluctuation information of the prediction results.

[0013] Furthermore, the identification of the adjustment demand according to the gas unit operation information and the fluctuation information of the prediction result includes: Obtain the gas units currently in operation and their corresponding outputs, and calculate the output deviation value of the gas units based on the fluctuation information of the prediction results; Calculate the output margin of the gas generator set currently in operation according to the gas generator set operation information; Based on the output margin and output deviation values, identify the need for adjustment of day-ahead planning and intraday scheduling.

[0014] Furthermore, the model parameters and cycle length of the multi-stage scheduling model are adjusted according to the identification result of the adjustment demand, combined with the operation information of the gas unit and the fluctuation information of the prediction result, including: When the output margin exceeds the output deviation value, the model parameters and cycle length of the day-ahead planning stage are maintained, and the output corresponding constraints of the intraday scheduling stage are adjusted according to the output deviation value and the operation information of the gas unit, and the cycle length of the intraday scheduling stage is shortened; When the output margin does not exceed the output deviation value, the start-up and shutdown scheduling cost parameters and the corresponding constraints of the gas units are adjusted in the day-ahead planning stage according to the operation information of the gas units, and the output corresponding constraints in the intraday scheduling stage are adjusted according to the output deviation value, and the cycle lengths of the day-ahead planning stage and the intraday scheduling stage are shortened.

[0015] Furthermore, the adjustment of the start-up and shutdown scheduling cost parameters and the corresponding constraints of the gas units in the day-ahead planning stage according to the gas unit operation information includes: Determine the operating time of the gas units currently in operation, and set the earliest start and stop control time of each gas unit according to the corresponding minimum continuous operating time and minimum shutdown time; The earliest start and stop control time of each gas unit is added to the corresponding constraint conditions of the gas unit.

[0016] Furthermore, the re-solving of the adjusted multi-stage adjustment model further includes: Every time the feedback node is reached, the prediction model for load conditions and renewable energy generation prediction is revised according to the fluctuation information of the prediction results; Re-forecast the load situation and renewable energy power generation based on the revised forecast model; Based on the re-obtained forecast results, the adjusted multi-stage adjustment model is solved.

[0017] The predicted fluctuations are determined by obtaining the actual operation and power generation information between adjacent feedback nodes, so as to accurately identify the adjustment needs, and then flexibly adjust the parameters and cycle length of the multi-stage scheduling model based on the relevant information. The day-ahead planning and intraday scheduling are further adjusted according to the output margin and deviation value. When the margin is sufficient, the intraday scheduling is accurately optimized. When the margin is insufficient, the planning and scheduling are fully adjusted to ensure the economy and reliability of the scheduling strategy. At the same time, the earliest start and stop control time of the gas unit is set and included in the constraint conditions to ensure the reasonable operation of the unit. Every time the feedback node is reached, the fluctuation information is used to correct the prediction model and re-predict, and the adjusted model is solved based on the new results, so that the scheduling strategy always keeps up with the actual changes, effectively improves the ability of the power system to cope with uncertainty, and achieves the multiple goals of cost reduction, efficient resource utilization and stable power supply.

[0018] Furthermore, solving the multi-stage scheduling model based on dual dynamic programming includes: For the day-ahead planning stage, define the corresponding dual dynamic planning stage according to the day-ahead planning cycle, and set the corresponding state variables; According to the corresponding constraints in the day-ahead planning stage, the corresponding dual variables are set and the corresponding dual objective function is constructed; Starting from the last stage, the optimal scheduling decision of this stage is calculated according to the values ​​of the current corresponding state variables and dual variables, and the optimal decision result of this stage is combined with the initial state of the next stage, moving forward to the previous stage until the optimal unit combination strategy of the day-ahead planning stage is obtained.

[0019] Furthermore, the solving of the multi-stage scheduling model based on dual dynamic programming also includes: For the intraday scheduling stage, the corresponding dual dynamic planning stage is defined according to the intraday planning cycle, and the corresponding state variables are set in combination with the optimal unit commitment strategy of the day-ahead planning stage; Set the corresponding dual variables according to the corresponding constraints in the intraday scheduling stage, and construct the corresponding dual objective function; Starting from the last stage, the optimal scheduling decision of this stage is calculated according to the values ​​of the current corresponding state variables and dual variables, and the optimal decision result of this stage is combined with the initial state of the next stage, moving forward to the previous stage until the optimal unit output allocation strategy for the intraday scheduling stage is obtained.

[0020] The beneficial effects of the present invention are: (1) Based on the rapid scheduling planning of gas-fired units by introducing pre-expanded multi-stage dual dynamic programming, the operation information of the gas-fired units and the predicted fluctuation information are used to verify and correct the load and renewable energy power generation forecast data by setting feedback nodes, thereby reducing the impact of data errors. This method dynamically adjusts the parameters and cycle length of the multi-stage scheduling model according to real-time information and re-solves and updates the scheduling strategy. This method can timely adjust the unit operation to ensure power supply, enhance the adaptability to extreme or complex change scenarios, further reduce the degree of dependence on data, reduce the impact of data errors through data verification and correction and model adaptive optimization, and improve the accuracy of scheduling planning.

[0021] (2) The feedback nodes are set by comprehensively considering multiple factors such as day-ahead planning information, intraday dispatch information, adjustment duration, time overlap, etc., and the feedback nodes are dynamically adjusted as the dispatch strategy is updated, providing data support that is more in line with real-time conditions for the dispatch decision of the gas unit, ensuring the stability and reliability of power supply. The dynamic update of feedback nodes can more timely obtain the real-time changes in the operating status of the gas unit, power load and renewable energy power generation. In the face of complex and changeable operating environments and emergencies such as extreme weather and large load fluctuations, it can quickly respond and adjust the dispatch strategy to effectively deal with various uncertainties. It can also capture the key time points of strategy adjustment in a timely manner, reduce unnecessary waiting and invalid operations, make the adjustment operation of the gas unit more efficient, and improve the overall operating efficiency of the power system.

[0022] (3) Determine the predicted fluctuation by obtaining the actual operation and power generation information between adjacent feedback nodes, so as to accurately identify the adjustment demand, and then flexibly adjust the parameters and cycle length of the multi-stage scheduling model based on the relevant information. Further, the day-ahead planning and intraday scheduling are adjusted according to the output margin and deviation value. When the margin is sufficient, the intraday scheduling is accurately optimized. When the margin is insufficient, the planning and scheduling are fully adjusted to ensure the economy and reliability of the scheduling strategy. At the same time, the earliest start-up and shutdown control time of the gas unit is set and included in the constraint conditions to ensure the reasonable operation of the unit. Every time the feedback node is reached, the prediction model is corrected and re-predicted using the fluctuation information, and the adjusted model is solved based on the new results, so that the scheduling strategy always keeps up with the actual changes, effectively improving the ability of the power system to cope with uncertainty, and achieving the multiple goals of cost reduction, efficient resource utilization and stable power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of a process of the present invention. DETAILED DESCRIPTION

[0024] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0025] Example: A gas unit scheduling method based on multi-stage dual dynamic programming, such as Figure 1 As shown, including: Predict the load conditions and renewable energy power generation in the power supply area corresponding to the gas-fired units; Taking cost minimization as the optimization goal and combining the forecast results, a multi-stage scheduling model for day-ahead unit combination planning and intra-day unit output scheduling in the power supply area is constructed; Solve the multi-stage scheduling model based on dual dynamic programming to obtain the scheduling strategy of the gas unit; Feedback nodes are set according to the day-ahead planning period and intraday scheduling period in the multi-stage scheduling model; When reaching the feedback node, obtain the gas unit operation information and the fluctuation information of the prediction results in the corresponding time period, and adjust the model parameters and cycle length of the multi-stage scheduling model accordingly; The adjusted multi-stage adjustment model is solved again and the dispatching strategy of the gas unit is updated until the dispatching of the gas unit for the day is completed.

[0026] Accurate load and renewable energy generation forecasts are the basis for subsequent dispatching decisions. Load forecasts are affected by many factors, such as season, time, weather conditions, and economic activities in the power supply area. Renewable energy generation, especially solar and wind power generation, is highly random and intermittent, and its power generation is closely related to meteorological conditions such as light intensity and wind speed.

[0027] On this basis, we first collect historical load data of the power supply area, new energy power generation data, related meteorological data and other influencing factors, clean the data, remove outliers and missing values, and normalize them for use in subsequent prediction models.

[0028] The load conditions and renewable energy power generation in the power supply area can be predicted by using traditional time series analysis methods, such as ARIMA (autoregressive integrated moving average model), or machine learning algorithms, such as neural networks, support vector machines and other prediction models, combined with processed historical load data, renewable energy power generation data and influencing factor data.

[0029] The established multi-stage scheduling model of day-ahead unit combination planning and intra-day unit output scheduling can be specifically divided into two stages, one is day-ahead unit combination planning, the other is intra-day unit output scheduling.

[0030] In the planning stage, the start-stop combination of gas units is determined in advance based on the load forecast and new energy power generation forecast results, and a preliminary framework is formulated for the subsequent power supply. Specifically, cost minimization is the core goal, and various cost factors are considered, such as gas cost, equipment maintenance cost, and start-stop cost.

[0031] Among them, the fuel cost is related to the amount of gas consumed when the gas unit generates electricity, and can be calculated based on the power generation efficiency of the gas unit and the gas price. The equipment maintenance cost is related to the operating time and start-stop times of the gas unit. Frequent start-stops will increase equipment wear and require more maintenance costs. The start-stop cost is the additional cost incurred each time the gas unit is started and stopped, including equipment preheating, cooling and other costs.

[0032] Therefore, the objective function of the day-ahead planning stage can be expressed as: ; in, is the estimated cost of the start-stop combination of the gas unit, is the number of gas units, is the number of time periods planned on the previous day, For the The fuel cost coefficient of each gas unit is For the The power generation of the gas-fired unit at time t, It is Maintenance cost coefficient of a gas unit, For the Gas units in The running time of the moment, is the start-stop cost coefficient, is the i-th gas unit in The start / stop status at the moment (start is 1, stop is 0).

[0033] The constraints in the day-ahead planning stage include power balance constraints, unit output constraints, and start-stop constraints.

[0034] Among them, the power balance constraint is that in each time period, the sum of the power generation of all gas units and the power generation of new energy must be equal to the load demand of that time period, and the unit output constraint is that the output of each gas unit should be between its minimum output and maximum output.

[0035] The start-stop constraints take into account the minimum continuous operating time and minimum continuous shutdown time of the unit to avoid frequent start-stop of the gas unit.

[0036] During the intraday dispatch stage, based on the day-ahead unit combination planning, the unit output is dynamically adjusted according to the real-time load changes during the day, the actual situation of renewable energy power generation, and the real-time operating status of the gas-fired units, in order to further optimize costs and ensure the stable operation of the power system.

[0037] The goal of intraday scheduling is still to minimize costs, but the real-time operation situation needs to be considered. In addition to fuel costs, maintenance costs, and start-up and shutdown costs, additional costs incurred by adjusting the unit output may also need to be considered, such as the loss cost of unit equipment when the load is increased or decreased rapidly.

[0038] Therefore, the objective function of this stage can be expressed as: ; in, is the number of time periods scheduled within the day, It is Cost coefficient for adjusting the output of gas generating units, For the i-th gas unit Output adjustment at any time.

[0039] The constraints in the intraday scheduling stage are similar to those in the day-ahead planning stage, including power balance constraints and unit output constraints. In addition, they also include ramp constraints for gas units.

[0040] Among them, the power balance constraints in the intraday scheduling stage are similar to those in the day-ahead planning stage, but the load data and new energy power generation used should use real-time data to ensure the balance of electricity supply and demand in each period.

[0041] Based on the above objective function and constraints, the multi-stage scheduling model is solved based on dual dynamic programming.

[0042] For the day-ahead planning stage, define the corresponding dual dynamic planning stage according to the day-ahead planning cycle, and set the corresponding state variables; According to the corresponding constraints in the day-ahead planning stage, the corresponding dual variables are set and the corresponding dual objective function is constructed; Starting from the last stage, the optimal scheduling decision of this stage is calculated according to the values ​​of the current corresponding state variables and dual variables, and the optimal decision result of this stage is combined with the initial state of the next stage, moving forward to the previous stage until the optimal unit combination strategy of the day-ahead planning stage is obtained.

[0043] When defining the stage, since the goal of day-ahead planning is generally one day, it can be divided into 24 time periods. Each time period can be defined as a stage of dual dynamic programming. Each stage corresponds to a time interval, and the decision on unit combination is made within this interval.

[0044] The state variables should contain key information that can describe the system state at the current stage in order to make reasonable scheduling decisions. In this embodiment, the start and stop status of the gas unit, the predicted value of load demand and the predicted value of new energy power generation are specifically set as state variables.

[0045] The corresponding dual variables are set according to the corresponding constraints in the day-ahead planning stage. On this basis, according to the objective function of the day-ahead planning, the constraints are introduced through Lagrangian relaxation to construct the corresponding dual objective function.

[0046] Starting from the last stage, the optimal decision of this stage is calculated through optimization algorithms such as linear programming. The optimal decision includes the start and stop status and power generation of each gas unit in the corresponding period of this stage under the goal of minimizing cost.

[0047] Then combine the optimal decision result of the last stage with the initial state of the previous stage, and use the optimization algorithm again to obtain the optimal decision of the corresponding stage.

[0048] The above forward recursive steps are repeated continuously, that is, the optimal decision of each stage is calculated and used as one of the initial conditions of the previous stage until the optimal decision of the first stage is calculated, so as to obtain the optimal start and stop status of each unit in each time period in the entire day-ahead planning stage, that is, the optimal unit combination strategy.

[0049] Similarly, for the intraday scheduling stage, the corresponding dual dynamic programming stage is defined according to the intraday planning cycle, and the corresponding state variables are set in combination with the optimal unit commitment strategy of the day-ahead planning stage; Set the corresponding dual variables according to the corresponding constraints in the intraday scheduling stage, and construct the corresponding dual objective function; Starting from the last stage, the optimal scheduling decision of this stage is calculated according to the values ​​of the current corresponding state variables and dual variables, and the optimal decision result of this stage is combined with the initial state of the next stage, moving forward to the previous stage until the optimal unit output allocation strategy for the intraday scheduling stage is obtained.

[0050] When defining the stages, although the goal of intraday scheduling is also one day, the time period is shorter. Specifically, 15 minutes can be used as a time period to implement the setting of the corresponding stage.

[0051] For intraday scheduling, the start and stop status of the gas unit, the actual value of renewable energy power generation, the actual value of load demand and the actual output of the gas unit are used as the state variables for intraday scheduling.

[0052] Under the corresponding stages and state variables, starting from the last stage, the algorithm is recursively forward to obtain the optimal output allocation strategy for each unit in each time period during the intraday scheduling stage.

[0053] Considering that the constraints and cycle lengths of the multi-stage scheduling model will be updated according to actual conditions, both the stage and state variables can be set to solve the optimal strategy, which can also adapt to the scenarios where the constraints and cycle lengths change.

[0054] After the multi-stage scheduling model is updated, when solving the multi-stage scheduling model based on dual dynamic programming, the executed time period is excluded, and the scheduling is re-planned according to the start and stop status and output of the gas units in the unfinished time period.

[0055] The setting of feedback nodes according to the day-ahead planning period and the intraday scheduling period in the multi-stage scheduling model includes: Determine the current day-ahead planning information and intra-day scheduling information according to the scheduling strategy; Taking the start time of each cycle as the candidate feedback node, the adjustment operation information of the gas unit at each candidate feedback node is calculated according to the corresponding day-ahead planning information and intraday scheduling information; Based on the adjustment operation information, the adjustment cost and adjustment duration are calculated at the alternative feedback nodes corresponding to the day-ahead planning and the alternative feedback nodes corresponding to the intraday scheduling respectively; Calculate the time overlap between the alternative feedback nodes corresponding to the day-ahead planning and intraday scheduling, and set the feedback nodes based on the adjustment cost and adjustment duration.

[0056] The day-ahead planning information includes the start-up and shutdown plan of the gas generator set, the power setting value for each period, etc., which is formulated on the day-ahead according to the load forecast and the new energy power generation forecast. The intraday dispatching information is the information related to the output planning of the gas generator set after the power generation of the unit is dynamically adjusted within the day based on the real-time load changes, the actual situation of new energy power generation and the real-time operation status of the gas generator set.

[0057] Taking into account the different cycle lengths used in the day-ahead planning and day-ahead scheduling stages, the candidate feedback nodes are selected for the day-ahead planning and day-ahead scheduling stages respectively. The cycle here can be the hourly period in the day-ahead planning, or a shorter time interval in the intraday scheduling, such as 15 minutes.

[0058] According to the day-ahead planning information and intraday dispatch information, the adjustment operation information of the gas unit at each alternative feedback node is calculated, including the start and stop operation of the unit, the adjustment amount of the power generation, etc. For example, in intraday dispatch, one of the alternative feedback nodes is 11 a.m. According to the load and renewable energy power generation at that time, as well as the previously formulated dispatch strategy, it can be calculated that one of the gas units needs to reduce the power generation from 150MW to 120MW at the alternative feedback node, where the power reduction operation is the adjustment operation information of the alternative feedback node.

[0059] Based on the adjustment operation information, the adjustment cost and adjustment duration of the alternative feedback nodes corresponding to the day-ahead planning and day-ahead scheduling are calculated.

[0060] Among them, the adjustment cost includes changes in fuel costs, equipment wear costs, etc. When the gas unit increases its power generation capacity, fuel consumption increases, and equipment wear will also increase accordingly. The adjustment cost is further calculated based on the wear costs corresponding to the increased fuel consumption and equipment wear.

[0061] The adjustment time refers to the time required to complete the corresponding adjustment operation. For example, at one of the alternative feedback nodes, the power generation capacity of one of the gas units needs to be increased from 100MW to 150MW. Due to the ramp rate limit of the gas unit, it takes 5 minutes to complete. These 5 minutes are the adjustment time.

[0062] For the time overlap between the alternative feedback nodes corresponding to the day-ahead planning and the intraday scheduling, the time overlap is determined by calculating the ratio of the overlapping time between the two to the total time. Before calculating the time overlap, the corresponding alternative feedback nodes are sorted according to the adjustment cost, adjustment time and execution time point. The higher the adjustment cost, the longer the adjustment time, the earlier the execution time point, and the earlier the final ranking. Then, the time overlap between the two alternative feedback nodes is calculated in sequence according to the ranking.

[0063] For the final feedback node, the alternative feedback nodes with higher adjustment cost, longer adjustment time and higher time overlap are preferred as the actual feedback nodes. High adjustment cost and long adjustment time often mean that the operation volume is large at this time node, and a small prediction deviation may cause a large execution deviation. Therefore, it is necessary to monitor the actual operation of the node and adjust the multi-stage scheduling model in a targeted manner to ensure that the multi-stage scheduling model is closer to reality and ensure the accuracy of subsequent scheduling strategies. A high time overlap can ensure that adjustments are made at key time points and timely capture time nodes with large deviations between the plan and the actual situation, thereby more effectively optimizing the scheduling strategy and ensuring the stable operation of the power system.

[0064] Each time the scheduling strategy is updated, the feedback nodes for the remaining time periods need to be re-evaluated because the parameters and operating conditions of the multi-stage scheduling model have changed.

[0065] Therefore, each time the scheduling strategy is updated, the remaining feedback nodes are reset according to the day-ahead planning period and intraday scheduling period in the adjusted multi-stage scheduling model.

[0066] After the feedback node is set, upon reaching the feedback node, the gas unit operation information and the fluctuation information of the prediction results within the corresponding time period are obtained, and the model parameters and cycle length of the multi-stage scheduling model are adjusted accordingly.

[0067] Specifically, the operation information of the gas unit, the actual load situation and the actual power generation information of the new energy in the time period from the previous feedback node to the current feedback node under the current scheduling strategy are obtained to determine the fluctuation information of the prediction result; Identify adjustment requirements based on gas unit operation information and fluctuation information of forecast results; According to the identification results of the adjustment demand, the model parameters and cycle length of the multi-stage scheduling model are adjusted in combination with the operation information of the gas units and the fluctuation information of the prediction results.

[0068] The previous feedback node mentioned is the time node of the previous execution of the multi-stage scheduling model.

[0069] The gas unit operation information includes the power generation power, start / stop status, equipment operation parameters, etc. of the gas unit.

[0070] After obtaining the actual load situation and the actual power generation information of renewable energy, they are compared with the predicted results to determine the fluctuation information of the predicted results.

[0071] Since the current dispatching strategy of gas turbines is based on the forecast results, the fluctuation information of the forecast results can intuitively reflect the execution deviation caused by the forecast deviation. If there is an execution deviation, there is a need for adjustment.

[0072] If the actual load situation is higher than the load situation in the forecast results, and the actual power generation information of new energy is consistent with the forecast results, if the gas-fired unit still operates according to the original scheduling strategy, the power generation capacity of the current power supply area will not be able to meet the load demand. This requires adjusting the unit start and stop strategy or adjusting the output of the existing operating units to compensate for this part of the load demand deviation.

[0073] Among them, the adjustment demand is identified based on the operation information of the gas unit and the fluctuation information of the forecast results, including: Obtain the gas units currently in operation and their corresponding outputs, and calculate the output deviation value of the gas units based on the fluctuation information of the prediction results; Calculate the output margin of the gas generator set currently in operation according to the gas generator set operation information; Based on the output margin and output deviation values, identify the need for adjustment of day-ahead planning and intraday scheduling.

[0074] In order to achieve accurate regulation, obtain the gas units currently in operation and the corresponding output, compare the output that should be provided under the fluctuation information of the predicted results with the actual output of all operating gas units, and determine the output deviation value of the gas unit that needs to be compensated.

[0075] The output margin refers to the range of power generation that can be increased or decreased under the current operating state of the gas unit. For each operating gas unit, the output margin can be calculated based on its technical parameters, such as maximum output and minimum output limits.

[0076] Compare the output margin and output deviation values ​​to identify the need for adjustment.

[0077] Among them, if the output deviation value is large and exceeds the output margin of the gas unit, it proves that the current scheduling strategy cannot meet the load demand, and it is necessary to adjust the unit start and stop plan or power generation power distribution planned in the day before.

[0078] If the output deviation value is within the output margin of the unit, the output of each gas unit needs to be adjusted. If the fluctuation of the predicted results persists or fluctuates greatly, the corresponding multi-stage scheduling model needs to be adjusted to compensate for this part of the execution deviation while better matching the real-time changes of load and renewable energy power generation.

[0079] Specifically, according to the identification result of the adjustment demand, the model parameters and cycle length of the multi-stage scheduling model are adjusted in combination with the operation information of the gas unit and the fluctuation information of the prediction result, including: When the output margin exceeds the output deviation value, the model parameters and cycle length of the day-ahead planning stage are maintained, and the output corresponding constraints of the intraday scheduling stage are adjusted according to the output deviation value and the operation information of the gas unit, and the cycle length of the intraday scheduling stage is shortened.

[0080] Since the output margin is sufficient to compensate for the output deviation, it means that the overall framework formulated in the current day-ahead planning stage can still meet the changes in power demand, so the model parameters and cycle length of the day-ahead planning stage are maintained. This means that key parameters such as the unit start-stop plan and power generation allocation formulated in the day-ahead planning do not need to be adjusted on a large scale, ensuring the stability and continuity of the day-ahead planning.

[0081] The output compensation target of the gas unit can be determined according to the output deviation value, and the operation information of the gas unit can provide restrictions for the adjustment of the constraint conditions. Specifically, the corresponding gas unit can be prompted to increase its output to meet the load demand by increasing the output lower limit constraint. At the same time, the real-time operation status of the gas unit can be determined in combination with the operation information of the gas unit. The output upper limit constraint can also be reasonably adjusted through the equipment temperature, pressure and other parameters of the gas unit to ensure that the gas unit increases its power generation capacity under the premise of safe operation.

[0082] The cycle length of the intraday scheduling phase will be further shortened to respond to real-time changes in load and renewable energy power generation in a more timely manner. According to the real-time output deviation value and the operation status of the gas unit, the unit output will be dynamically adjusted to achieve a more accurate balance between electricity supply and demand.

[0083] When the output margin does not exceed the output deviation value, the start-up and shutdown scheduling cost parameters and the corresponding constraints of the gas units are adjusted in the day-ahead planning stage according to the operation information of the gas units, and the output corresponding constraints in the intraday scheduling stage are adjusted according to the output deviation value, and the cycle lengths of the day-ahead planning stage and the intraday scheduling stage are shortened.

[0084] When the output margin is insufficient, more gas units may need to be started to make up for the output deviation.

[0085] At this time, for the day-ahead planning stage, the start-stop scheduling cost parameters are adjusted to optimize the start-stop decision of the gas unit, such as reducing the start-stop cost coefficient of the newly started gas unit, so that more gas units can be started when necessary to meet the power demand. At the same time, considering the operating efficiency and cost differences of different gas units, the fuel cost coefficient of each gas unit is also adjusted accordingly to keep the overall power generation cost within a controllable range.

[0086] At the same time, the output restrictions of some gas-fired units will be relaxed, and gas-fired units will be allowed to operate within a wider power range under the premise of ensuring equipment safety. At the same time, the constraints such as the minimum continuous operation time and the minimum continuous shutdown time of gas-fired units will also be appropriately adjusted to meet the power supply needs in emergency situations.

[0087] For the intraday scheduling stage, similar to when the output margin exceeds the output deviation value, the output corresponding constraint conditions of the intraday scheduling stage are adjusted according to the output deviation value.

[0088] At the same time, the cycle lengths of the day-ahead planning stage and the intraday dispatching stage are shortened. For the day-ahead planning stage, the cycle length originally in hours can be appropriately shortened to allow for more frequent evaluation and adjustment of the start-stop plan and power distribution of the units. The intraday dispatching stage will further shorten the cycle, such as from 10 minutes to 5 minutes, to achieve rapid response to real-time changes in the power system, and through close monitoring and timely adjustments, try to meet power demand and ensure the stable operation of the power system.

[0089] However, in addition to the above situations, there are also cases where the output deviation value is caused by the inaccurate prediction of the output of new energy sources, resulting in excessive power generation of the gas-fired unit. For such cases, the output deviation value and the output margin can also be compared to adjust the day-ahead planning stage or the intraday planning stage to achieve processing. The output margin here means that the gas-fired unit can reduce its output.

[0090] The adjustment of the start-up and shutdown scheduling cost parameters and the corresponding constraints of the gas units in the day-ahead planning stage according to the gas unit operation information includes: Determine the operating time of the gas units currently in operation, and set the earliest start and stop control time of each gas unit according to the corresponding minimum continuous operating time and minimum shutdown time; The earliest start and stop control time of each gas unit is added to the corresponding constraint conditions of the gas unit.

[0091] The operating time of the gas unit currently in operation is directly related to whether the gas unit can be started and stopped according to the requirements of the minimum continuous operating time and the minimum shutdown time. Assuming that the minimum continuous operating time of a gas unit is 4 hours, if the unit has been running for 3 hours, then it cannot be shut down in the next 1 hour, that is, the earliest shutdown time must be at least 1 hour later. Similarly, if a gas unit has just been shut down and its minimum shutdown time is 2 hours, then it cannot be started again within 2 hours, and the earliest startup time is 2 hours after the shutdown. Through the above calculations, the earliest startup and earliest shutdown time points under the current operating state can be determined for each gas unit.

[0092] The earliest start-up and earliest shutdown time points for each gas-fired unit under the current operating status are determined as constraints and added to the day-ahead planning stage to avoid unreasonable scheduling that violates the minimum continuous operating time and minimum downtime requirements, ensuring the stable operation and equipment life of the gas-fired units, while also improving the reliability and economy of the entire power system.

[0093] The re-solving of the adjusted multi-stage adjustment model further includes: Every time the feedback node is reached, the prediction model for load conditions and renewable energy generation prediction is revised according to the fluctuation information of the prediction results; Re-forecast the load situation and renewable energy power generation based on the revised forecast model; Based on the re-obtained forecast results, the adjusted multi-stage adjustment model is solved.

[0094] Taking into account that the initial training of the prediction model is based on historical data, the prediction results are inevitably biased, and the fluctuation value of the prediction results can be intuitively obtained at the feedback node. The latest actual load data, new energy power generation data and corresponding meteorological data are used as new training samples at the feedback node to retrain the prediction model, adjust the model's weights and parameters, and improve the model's prediction accuracy for future loads and new energy power generation, so as to further improve the accuracy of the scheduling decision of the multi-stage scheduling model.

[0095] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A gas unit scheduling method based on multi-stage dual dynamic programming, characterized in that: include: Predict the load conditions and renewable energy power generation in the power supply area corresponding to the gas-fired units; Taking cost minimization as the optimization goal and combining the forecast results, a multi-stage scheduling model for day-ahead unit combination planning and intra-day unit output scheduling in the power supply area is constructed; Solve the multi-stage scheduling model based on dual dynamic programming to obtain the scheduling strategy of the gas unit; Feedback nodes are set according to the day-ahead planning period and intraday scheduling period in the multi-stage scheduling model; When reaching the feedback node, obtain the gas unit operation information and the fluctuation information of the prediction results in the corresponding time period, and adjust the model parameters and cycle length of the multi-stage scheduling model accordingly; The adjusted multi-stage adjustment model is solved again and the dispatching strategy of the gas unit is updated until the dispatching of the gas unit for the day is completed.

2. The method for dispatching gas units based on multi-stage dual dynamic programming according to claim 1 is characterized in that: The setting of feedback nodes according to the day-ahead planning period and the intraday scheduling period in the multi-stage scheduling model includes: Determine the current day-ahead planning information and intra-day scheduling information according to the scheduling strategy; Taking the start time of each cycle as the candidate feedback node, the adjustment operation information of the gas unit at each candidate feedback node is calculated according to the corresponding day-ahead planning information and intraday scheduling information; Based on the adjustment operation information, the adjustment cost and adjustment duration are calculated at the alternative feedback nodes corresponding to the day-ahead planning and the alternative feedback nodes corresponding to the intraday scheduling respectively; Calculate the time overlap between the alternative feedback nodes corresponding to the day-ahead planning and intraday scheduling, and set the feedback nodes based on the adjustment cost and adjustment duration.

3. The method for dispatching gas turbine units based on multi-stage dual dynamic programming according to claim 2 is characterized in that: The feedback node is set according to the day-ahead planning period and the intra-day scheduling period in the multi-stage scheduling model, and further includes: Each time the scheduling strategy is updated, the remaining feedback nodes are reset according to the day-ahead planning period and intraday scheduling period in the adjusted multi-stage scheduling model.

4. The method for dispatching gas units based on multi-stage dual dynamic programming according to claim 1, characterized in that: When the feedback node is reached, the operation information of the gas unit and the fluctuation information of the prediction result in the corresponding time period are obtained, and the model parameters and cycle length of the multi-stage scheduling model are adjusted accordingly, including: Obtain the gas unit operation information, actual load conditions, and actual power generation information of new energy sources in the time period from the previous feedback node to the current feedback node under the current scheduling strategy, and determine the fluctuation information of the prediction results; Identify adjustment requirements based on gas unit operation information and fluctuation information of forecast results; According to the identification results of the adjustment demand, the model parameters and cycle length of the multi-stage scheduling model are adjusted in combination with the operation information of the gas units and the fluctuation information of the prediction results.

5. The method for dispatching gas turbine units based on multi-stage dual dynamic programming according to claim 4 is characterized in that: The identifying of the adjustment demand according to the gas unit operation information and the fluctuation information of the prediction result includes: Obtain the gas units currently in operation and their corresponding outputs, and calculate the output deviation value of the gas units based on the fluctuation information of the prediction results; Calculate the output margin of the gas generator set currently in operation according to the gas generator set operation information; Based on the output margin and output deviation values, identify the need for adjustment of day-ahead planning and intraday scheduling.

6. The method for dispatching gas turbine units based on multi-stage dual dynamic programming according to claim 5, characterized in that: The method of adjusting the model parameters and cycle length of the multi-stage scheduling model according to the identification result of the adjustment demand and combining the operation information of the gas unit and the fluctuation information of the prediction result includes: When the output margin exceeds the output deviation value, the model parameters and cycle length of the day-ahead planning stage are maintained, and the output corresponding constraints of the intraday scheduling stage are adjusted according to the output deviation value and the operation information of the gas unit, and the cycle length of the intraday scheduling stage is shortened; When the output margin does not exceed the output deviation value, the start-up and shutdown scheduling cost parameters and the corresponding constraints of the gas units are adjusted in the day-ahead planning stage according to the operation information of the gas units, and the output corresponding constraints in the intraday scheduling stage are adjusted according to the output deviation value, and the cycle lengths of the day-ahead planning stage and the intraday scheduling stage are shortened.

7. The method for dispatching gas turbine units based on multi-stage dual dynamic programming according to claim 6, characterized in that: The adjustment of the start-up and shutdown scheduling cost parameters and the corresponding constraints of the gas units in the day-ahead planning stage according to the gas unit operation information includes: Determine the operating time of the gas units currently in operation, and set the earliest start and stop control time of each gas unit according to the corresponding minimum continuous operating time and minimum shutdown time; The earliest start and stop control time of each gas unit is added to the corresponding constraint conditions of the gas unit.

8. The method for dispatching gas turbine units based on multi-stage dual dynamic programming according to claim 4, characterized in that: The re-solving of the adjusted multi-stage adjustment model further includes: Every time the feedback node is reached, the prediction model for load conditions and renewable energy generation prediction is revised according to the fluctuation information of the prediction results; Re-forecast the load situation and renewable energy power generation based on the revised forecast model; Based on the re-obtained forecast results, the adjusted multi-stage adjustment model is solved.

9. The method for dispatching gas units based on multi-stage dual dynamic programming according to claim 1, characterized in that: The method of solving the multi-stage scheduling model based on dual dynamic programming includes: For the day-ahead planning stage, define the corresponding dual dynamic planning stage according to the day-ahead planning cycle, and set the corresponding state variables; According to the corresponding constraints in the day-ahead planning stage, the corresponding dual variables are set and the corresponding dual objective function is constructed; Starting from the last stage, the optimal scheduling decision of this stage is calculated according to the values ​​of the current corresponding state variables and dual variables, and the optimal decision result of this stage is combined with the initial state of the next stage, moving forward to the previous stage until the optimal unit combination strategy of the day-ahead planning stage is obtained.

10. The method for dispatching gas turbine units based on multi-stage dual dynamic programming according to claim 1, characterized in that: The solving of the multi-stage scheduling model based on dual dynamic programming also includes: For the intraday scheduling stage, the corresponding dual dynamic planning stage is defined according to the intraday planning cycle, and the corresponding state variables are set in combination with the optimal unit commitment strategy of the day-ahead planning stage; Set the corresponding dual variables according to the corresponding constraints in the intraday scheduling stage, and construct the corresponding dual objective function; Starting from the last stage, the optimal scheduling decision of this stage is calculated according to the values ​​of the current corresponding state variables and dual variables, and the optimal decision result of this stage is combined with the initial state of the next stage, moving forward to the previous stage until the optimal unit output allocation strategy for the intraday scheduling stage is obtained.

Images