Multi-target collaborative optimization operation method and system for gas generator set

Abstract

The invention discloses a multi-target collaborative optimization operation method and system for a gas generator set, and relates to the technical field of gas generator set control. A noise superposition model is adopted to calculate the total superposition noise sound pressure level of each noise sensitive point, and meanwhile, a final collaborative operation scheme is selected according to preset preference; the load-noise dynamic characteristics of the unit are comprehensively considered, through the modular architecture and micro-service design, the environmental noise limit values of different noise sensitive areas can be quickly adapted, the power supply reliability can be ensured, adjustment can be performed according to different operation conditions, load requirements and environmental requirements, and the power supply efficiency is improved. It is ensured that the flexible coping capacity is kept in a continuously changing market environment, joint optimization of a noise target and an electric power cost target is achieved, and the limitation of traditional discrete control is broken through.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas generator set control, and particularly relates to a multi-objective collaborative optimization operation method and system for a gas generator set. Background Art

[0002] With the acceleration of the urbanization and industrialization processes, gas generator sets play an important role in distributed energy systems due to their high efficiency and flexibility. However, when multiple gas generator sets are deployed in noise-sensitive areas such as residential areas, hospitals, and schools, the noise superposition effect generated during their operation has become a key issue restricting equipment deployment and compliance operation. In the prior art, the noise control and operation optimization of gas generator sets mainly rely on single-objective control strategies or static noise reduction means. The noise prediction model and the power dispatching system adopt a discrete architecture, which generally mainly focuses on the noise level detection of a single device, lacking a systematic evaluation of the mutual influence between noise sources during the combined operation of multiple devices. The data interaction delay leads to lagging control instructions and a lack of dynamic collaborative optimization ability, making it difficult to achieve multi-objective collaborative optimization of noise and operation cost under complex working conditions. Summary of the Invention

[0003] Aiming at the defects in the prior art, the present invention provides a multi-objective collaborative optimization operation method and system for a gas generator set to solve the above technical problems.

[0004] On the one hand, a multi-objective collaborative optimization operation method for a gas generator set is provided, including the following contents: Determine the environmental noise limit values of each noise-sensitive point, and obtain the noise source characteristic data of each generator set and its position information relative to the noise-sensitive point; Based on the noise source characteristic data and position information of each gas generator set, establish a noise superposition model, which is used to calculate the total superposition noise sound pressure level of each noise-sensitive point under any operating combination of gas generator sets; Taking the output power of each gas generator set as a decision variable, construct a multi-objective function with the optimization objectives of minimizing the total superposition noise of key noise-sensitive points and minimizing the total operation cost, use the target genetic algorithm to solve the Pareto optimal solution set, and select the final collaborative operation plan according to the preset preference.

[0005] Preferably, when establishing the noise superposition model based on the noise source characteristic data and position information of each gas generator set, the following steps are specifically included: Use the point cloud registration algorithm to align the position information of different gas generator sets, and the position information includes the terrain elevation data and obstacle distribution on the noise propagation path of each generator set; Based on the noise source characteristic data, obtain the noise-power relationship curve of the gas generator set under different load rates; Establish a noise superposition model by combining the three-dimensional coordinates of noise-sensitive points.

[0006] Preferably, the noise superposition model is specifically as follows: , where is the total sound pressure level of the sensitive point , is the sound pressure level of the th unit at the sensitive point ; [[ID=2,1]] where is the sound power level of the gas generator set at the output power , is the equivalent noise reduction amount, is the straight-line distance between the gas generator set and the sensitive point , is the reference distance, is the atmospheric absorption coefficient, is the ambient temperature of the environment where the gas generator set is located, is the ambient humidity of the environment where the gas generator set is located.

[0007] Preferably, the specific calibration method of the noise-power relationship curve is as follows: Within the operating load range of the unit, measure the octave band noise spectrum under each working condition at a set interval; Generate a continuous function based on the obtained octave band noise spectrum through cubic spline interpolation.

[0008] Preferably, when generating a continuous function based on the obtained octave band noise spectrum through cubic spline interpolation, it specifically includes the following steps: Normalize the operating load data, and use the adaptive node distribution algorithm to increase the interpolation node density in the interval with large load gradient changes; Based on the Z-score standardization of the octave band noise spectrum, force the setting of nodes for the mutation points generated after standardization.

[0009] Preferably, taking the output power of each gas generator set as the decision variable, when constructing a multi-objective function with minimizing the total superimposed noise of key noise-sensitive points and minimizing the total operating cost as the optimization objectives and using the target genetic algorithm to solve the Pareto optimal solution set, it specifically includes the following steps: Based on the constructed multi-objective function, divide the stages according to the noise-sensitive periods during the execution of the power generation plan, obtain the noise limits for each stage, construct the standard noise curve, and set the stage noise upper limit in combination with the environmental noise limit. Generate the stage noise distribution curve corresponding to the current operating combination of gas-fired generator sets based on the noise superposition model, and dynamically compare it with the standard noise curve. Based on the comparison results, use the multi-objective genetic algorithm to solve the Pareto optimal solution set.

[0010] Preferably, when obtaining the noise limits for each stage and constructing the standard noise curve, it specifically includes the following steps: Based on the historical noise monitoring data and the operating logs of gas-fired generator sets, divide the peak noise stage and the trough noise stage according to the time dimension. Use the equivalent noise energy method to calculate the allowable noise reference values for the peak noise stage and the trough noise stage respectively, and integrate the noise reference values to obtain the standard noise curve.

[0011] Preferably, the preset preference is realized through at least one of the following methods: Set the priority weights of noise-sensitive points and perform weighted sorting on the Pareto front solutions. Select the solution with the lowest noise under the condition that the increase in the total operating cost does not exceed the set constraint. Manually and interactively select the solution that meets the noise reduction requirements for the set time period.

[0012] On the other hand, provide a multi-objective collaborative optimization operation system for gas-fired generator sets, including the following: Geographical acquisition module: The geographical acquisition module includes means for acquiring the location information of gas-fired generator sets and constructing a three-dimensional propagation path model based on the acquired location information. Noise monitoring network: The noise monitoring network is a distributed acoustic sensor array deployed at each sensitive point. Acoustic calculation module: The acoustic calculation module integrates the noise superposition model and is used to calculate the total superimposed noise sound pressure level at each noise-sensitive point under any operating combination of generator sets. Coupled optimization module: The coupled optimization module integrates the target genetic algorithm and is used to solve the Pareto optimal solution set. Multi-objective decision-making module: The multi-objective decision-making module selects the final collaborative operation plan based on the Pareto optimal solution set according to the preset preference.

[0013] Preferably, it further includes a multi-source heterogeneous data fusion module and a dynamic scheduling execution unit. The multi-source heterogeneous data fusion module is used to integrate the location information and the output power. The dynamic scheduling execution unit is used to control the power output and start-stop timing sequence of gas-fired generator sets according to the final collaborative operation plan.

[0014] The beneficial effects of the present invention are as follows: The present invention uses a noise superposition model to calculate the total superposition noise sound pressure level of each noise-sensitive point, and at the same time selects the final collaborative operation plan according to the preset preference, comprehensively considering the unit load-noise dynamic characteristics. Through the modular architecture and microservice design, it can not only quickly adapt to the environmental noise limits of different noise-sensitive areas, but also ensure power supply reliability, and can be adjusted according to different operating conditions, load demands and environmental requirements, ensuring the ability to respond flexibly in the changing market environment, realizing the joint optimization of noise objectives and power cost objectives, and breaking through the limitations of traditional discrete control. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a step flow chart of a multi-objective collaborative optimization operation method for a gas generator set provided by the present invention; Figure 2 It is a schematic structural diagram of a multi-objective collaborative optimization operation system for a gas generator set provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0018] The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are only examples and are not intended to limit the present invention.

[0019] The embodiments of the invention will be described in detail below with reference to the drawings.

[0020] As Figure 1As shown in the figure, a multi-objective collaborative optimization operation method for a gas generator set includes the following contents: Determine the environmental noise limit values of each noise-sensitive point, and obtain the noise source characteristic data of each generator set and its position information relative to the noise-sensitive point; Based on the noise source characteristic data and position information of each gas generator set, establish a noise superposition model, which is used to calculate the total superposition noise sound pressure level of each noise-sensitive point under any operating combination of gas generator sets; Taking the output power of each gas generator set as a decision variable, construct a multi-objective function with the optimization objectives of minimizing the total superposition noise of key noise-sensitive points and minimizing the total operating cost, use the target genetic algorithm to solve the Pareto optimal solution set, and select the final collaborative operation plan according to the preset preference.

[0021] The present invention uses a noise superposition model to calculate the total superposition noise sound pressure level of each noise-sensitive point, and at the same time selects the final collaborative operation plan according to the preset preference, comprehensively considers the unit load-noise dynamic characteristics, and through the modular architecture and microservice design, can not only quickly adapt to the environmental noise limit values of different noise-sensitive areas, but also ensure the power supply reliability, and can be adjusted according to different operating conditions, load demands and environmental requirements, ensuring the ability to respond flexibly in the changing market environment, realizing the joint optimization of noise objectives and power cost objectives, and breaking through the limitations of traditional discrete control.

[0022] More specifically, when establishing a noise superposition model based on the noise source characteristic data and position information of each gas generator set, it specifically includes the following steps: Use the point cloud registration algorithm to align the position information of different gas generator sets, and the position information includes the terrain elevation data and obstacle distribution on the noise propagation path of each generator set; Based on the noise source characteristic data, obtain the noise-power relationship curve of the gas generator set under different load rates; Combine the three-dimensional coordinates of the noise-sensitive point to establish a noise superposition model.

[0023] By using the point cloud registration algorithm to align the three-dimensional position information of multiple units and integrating the terrain elevation and obstacle distribution data, the calculation error problem of sound wave diffraction / reflection caused by traditional manual measurement or two-dimensional simplified modeling is solved. At the same time, the noise source characteristic data includes the sound power level spectrum under rated conditions. Based on the measured noise-power relationship curve, a sound source model is constructed on the basis of the sound wave propagation theory, which can accurately reflect the noise spectrum change characteristics of the unit under different load rates, effectively quantify the obstacle shielding effect, and is especially suitable for complex acoustic environments such as urban dense areas and mountains.

[0024] More specifically, the noise superposition model is specifically as follows: , wherein, is the total sound pressure level of the sensitive point, and is the sound pressure level of the th unit at the sensitive point ; , wherein, is the sound power level of the gas generator set at the output power , is the equivalent noise reduction amount, is the straight-line distance between the gas generator set [[ID=2 / 28]] and the sensitive point , is the reference distance, is the atmospheric absorption coefficient, is the ambient temperature of the environment where the gas generator set is located, is the ambient humidity of the environment where the gas generator set is located.

[0025] Among them, the sound power level is obtained by the following formula: , is the unit characteristic coefficient of the gas generator set , is the sound power level of the gas generator set at the rated power .

[0026] Among them, the atmospheric absorption coefficient is obtained by the following formula: , wherein, is the noise center frequency, and the equivalent noise reduction amount is calculated according to ISO 9613-2 or directly input.

[0027] When calculating the total superimposed noise sound pressure level of the noise sensitive point, the following formula is specifically used: When calculating the total sound pressure level, the following constraints also need to be introduced: Unit power constraint: , wherein, is the maximum output power of the gas generator set , is the minimum output power of the gas generator set .

[0028] Safety distance constraint: , wherein, is the gas generator set Distance to noise-sensitive point The safety distance.

[0029] More specifically, the specific calibration method of the noise-power relationship curve is as follows: Within the operating load range of the unit, measure the octave band noise spectrum under each working condition at a set interval; Based on the obtained octave band noise spectrum, generate a continuous function by cubic spline interpolation method.

[0030] Among them, the test range is 30%-100% of the unit's load range, the set interval is 5%, and the octave band noise spectrum is 1 / 3 of the range under each working condition.

[0031] Dense sampling is carried out at 5% intervals within the 30%-100% load range of the unit. Compared with the traditional sampling method with a 10%-15% interval, it can capture the non-linear influence of the load rate change on the noise spectrum, such as the noise mutation caused by combustion instability in the 50%-60% load interval. And by using the cubic spline interpolation method instead of the linear interpolation method, it can ensure that the noise-power relationship curve is second-order continuously differentiable, eliminate the curvature jump problem at the test points of the traditional sampling method, and compared with the traditional 1 / 1 octave band or A-weighted total sound level measurement, it can accurately identify the load sensitivity of specific frequency bands such as 125Hz low-frequency noise, providing data support for targeted noise reduction.

[0032] More specifically, when generating a continuous function by cubic spline interpolation method based on the obtained octave band noise spectrum, it specifically includes the following steps: Normalize the working load data, and use the adaptive node distribution algorithm to increase the interpolation node density in the interval with large load gradient changes; Based on the Z-score standardization of the octave band noise spectrum, force nodes to be set for the mutation points generated after standardization.

[0033] When normalizing the working load data, the following formula is specifically used: , Among them, is the measured discrete load point in the working load data, is the upper limit of the load test range, is the lower limit of the load test range.

[0034] Through load normalization and adaptive node control, the number of redundant nodes can be reduced on the premise of ensuring accuracy, avoiding the over-smoothing problem of traditional cubic spline interpolation for data mutation, improving the feature retention rate. At the same time, Z-score standardization can eliminate the dimension difference of sound power levels in different frequency bands, making the interpolation weight distribution more reasonable.

[0035] More specifically, taking the output power of each gas-fired generator set as the decision variable, a multi-objective function with the optimization objectives of minimizing the total superimposed noise at key noise-sensitive points and minimizing the total operating cost is constructed. When using the objective genetic algorithm to solve the Pareto optimal solution set, the following steps are specifically included: Based on the constructed multi-objective function, divide the stages according to the noise-sensitive periods during the implementation of the power generation plan, obtain the noise limits for each stage and construct a standard noise curve, and set the stage noise upper limit in combination with the environmental noise limit; Generate the stage noise distribution curve corresponding to the current operating combination of gas-fired generator sets based on the noise superposition model, and dynamically compare it with the standard noise curve; Based on the comparison results, use the multi-objective genetic algorithm to solve the Pareto optimal solution set.

[0036] After using the improved multi-objective optimization algorithm to solve the Pareto optimal solution set, based on the pre-established mapping database of unit noise-power-efficiency, filter out the high-efficiency and low-noise operating intervals that meet and , wherein, is the noise generated by the gas-fired generator set at the output power , is the noise limit, is the gas-fired generator set at the output power when the working efficiency, is the lower limit of the working efficiency of the gas-fired generator set; During the optimization process, low-noise units are preferentially allocated to undertake the base load, so that high-noise units are only used for peak load regulation.

[0037] By dividing the noise-sensitive periods and setting the dynamic noise upper limit, compared with the traditional all-day fixed limit mode, while ensuring noise reduction during key periods, moderate noise is allowed during non-sensitive periods to reduce the operating cost, and the construction of the standard noise curve makes the optimization objective more in line with the actual regulatory requirements, avoiding the excessive conservatism of the traditional root mean square index.

[0038] More specifically, when obtaining the noise limits for each stage and constructing the standard noise curve, the following steps are specifically included: Based on the historical noise monitoring data and the operation logs of gas-fired generator sets, divide the peak noise stage and the trough noise stage according to the time dimension; Respectively use the equivalent noise energy method to calculate the allowable noise reference values for the peak noise stage and the trough noise stage, and integrate the noise reference values to obtain the standard noise curve.

[0039] Divide the peak / trough noise stage through historical data. For example, the peak period in the industrial park is from 8:00 to 18:00 during the day, and the trough period at night is from 22:00 to 6:00. Compared with the traditional unified limit mode throughout the day, noise control is more in line with the actual needs. At the same time, based on the unit operation log, obtain data such as start / stop time and load rate, and correlate with the noise stage division to dynamically adjust the standard curve, and special high-noise events such as temporary equipment maintenance periods can be identified.

[0040] More specifically, the preset preference is achieved through at least one of the following methods: Set the priority weights of noise-sensitive points and perform weighted sorting on the Pareto front solutions; Select the solution with the lowest noise under the condition that the total operating cost increase does not exceed the set constraint; Manually and interactively select the solution that meets the noise reduction requirements during the set time period.

[0041] The priority weight mechanism sets different sensitivity point weights through different limit values of noise-sensitive points, making the sorting of the Pareto solution set more in line with the actual management needs. Compared with the traditional equal weight method, the noise reduction amplitude in the key area is increased by 20%-30%. The cost constraint-oriented selection automatically screens the lowest noise solution under the hard constraint that the total operating cost increase ≤ 5%, solving the inefficiency problem of manual traversal of the solution set in the traditional scheme; Manually and interactively selecting allows adjusting the solution set in combination with the real-time scenario, improving the operability of the scheme.

[0042] A multi-objective collaborative optimization operation system for gas generator sets includes the following: Geographical acquisition module: The geographical acquisition module includes means for acquiring the location information of the gas generator set and constructing a three-dimensional propagation path model based on the acquired location information; Noise monitoring network: The noise monitoring network is a distributed acoustic sensor array deployed at each sensitive point; Acoustic calculation module: The acoustic calculation module integrates a noise superposition model for calculating the total superimposed noise sound pressure level at each noise-sensitive point under any generator set operation combination; Coupled optimization module: The coupled optimization module integrates a target genetic algorithm for solving the Pareto optimal solution set; Multi-objective decision-making module: The multi-objective decision-making module selects the final collaborative operation plan based on the Pareto optimal solution set according to the preset preference.

[0043] As Figure 2 shown, more specifically, it also includes a multi-source heterogeneous data fusion module and a dynamic scheduling execution unit. The multi-source heterogeneous data fusion module is used to integrate location information and output power; The dynamic scheduling execution unit is used to control the power output and start / stop timing sequence of the gas generator set according to the final co - operation operation plan.

[0044] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and the description of the present invention.

Claims

1. A multi-objective collaborative optimization operation method for a gas generator set, characterized in that It includes the following steps: Determine the environmental noise limits of each noise-sensitive point, and obtain the noise source characteristic data of each generator set and its position information relative to the noise-sensitive point; Based on the noise source characteristic data and position information of each gas generator set, establish a noise superposition model, which is used to calculate the total superposition noise sound pressure level of each noise-sensitive point under any operating combination of gas generator sets; Taking the output power of each gas generator set as the decision variable, construct a multi-objective function with the optimization objectives of minimizing the total superposition noise of key noise-sensitive points and minimizing the total operating cost, use the target genetic algorithm to solve the Pareto optimal solution set, and select the final co-operation operation plan according to the preset preference.

2. The multi-objective collaborative optimization operation method of a gas-fired generating set according to claim 1, characterized in that When establishing the noise superposition model based on the noise source characteristic data and position information of each gas generator set, it specifically includes the following steps: Use the point cloud registration algorithm to align the position information of different gas generator sets. The position information includes the terrain elevation data and obstacle distribution on the noise propagation path of each generator set; Based on the noise source characteristic data, obtain the noise-power relationship curve of the gas generator set under different load rates; Combine the three-dimensional coordinates of the noise-sensitive point to establish a noise superposition model.

3. The multi-objective collaborative optimization operation method of a gas generator set according to claim 2, characterized in that, The noise superposition model is specifically as follows: , Among them, is the sensitive point of the total sound pressure level, is the th unit's sound pressure level at the sensitive point ; ， Among them, is the gas generator set at the output power of the sound power level, is the equivalent noise reduction amount, is the gas generator set and the sensitive point of the straight-line distance, is the reference distance, is the atmospheric absorption coefficient, is the gas generator set where the ambient temperature of the environment, is the gas generator set where the ambient humidity of the environment.

4. The multi-objective collaborative optimization operation method of a gas generator set according to claim 2, characterized in that, The specific calibration method of the noise-power relationship curve is as follows: In different working load ranges of the unit, measure the octave band noise spectrum under each working condition at the set interval; Based on the obtained octave band noise spectrum, generate a continuous function by cubic spline interpolation method.

5. The multi-objective collaborative optimization operation method of a gas-fired power generation unit according to claim 4, characterized in that When generating a continuous function by cubic spline interpolation method based on the obtained octave band noise spectrum, it specifically includes the following steps: Normalize the working load data, and use the adaptive node distribution algorithm to increase the interpolation node density in the interval with large load gradient change; Based on the Z-score standardization of the octave band noise spectrum, force the setting of nodes for the mutation points generated after standardization.

6. The multi-objective collaborative optimization operation method of a gas generator set according to claim 1, characterized in that, When taking the output power of each gas generator set as the decision variable and constructing a multi-objective function with the optimization objectives of minimizing the total superposition noise of key noise-sensitive points and minimizing the total operating cost, and using the target genetic algorithm to solve the Pareto optimal solution set, it specifically includes the following steps: Based on the constructed multi-objective function, divide the stages according to the noise-sensitive time periods during the execution of the power generation plan, obtain the noise limits of each stage and construct a standard noise curve, and set the stage noise upper limit in combination with the environmental noise limit; Generate the stage noise distribution curve corresponding to the current operating combination of gas generator sets based on the noise superposition model, and make a dynamic comparison with the standard noise curve; Based on the comparison result, use the multi-objective genetic algorithm to solve the Pareto optimal solution set.

7. The multi-objective collaborative optimization operation method of a gas generating set according to claim 6, characterized in that When obtaining the noise limits of each stage and constructing a standard noise curve, it specifically includes the following steps: Based on the historical noise monitoring data and the operation log of the gas generator set, divide the peak noise stage and the valley noise stage according to the time dimension; Use the equivalent noise energy method to calculate the allowable noise reference value for the peak noise stage and the valley noise stage respectively, and integrate the noise reference values to obtain the standard noise curve.

8. The multi-objective collaborative optimization operation method of a gas generator set according to claim 1, characterized in that The preset preference is realized through at least one of the following methods: Set the priority weights of the noise-sensitive points, and perform weighted sorting on the Pareto front solutions; Select the solution with the lowest noise level under the condition that the increase in the total operating cost does not exceed the set constraint; Manually and interactively select the solution that meets the noise reduction requirements during the set period.

9. A multi-objective collaborative optimization operation system for a gas-fired generator set, characterized in that, It includes the following contents: Geographical acquisition module: The geographical acquisition module includes means for acquiring the position information of the gas generator sets and constructing a three-dimensional propagation path model based on the acquired position information; Noise monitoring network: The noise monitoring network is a distributed acoustic sensor array deployed at each sensitive point; Acoustic calculation module: The acoustic calculation module integrates a noise superposition model for calculating the total superimposed noise sound pressure level at each noise-sensitive point under any operating combination of the generator sets; Coupled optimization module: The coupled optimization module integrates a target genetic algorithm for solving the Pareto optimal solution set; Multi-objective decision-making module: The multi-objective decision-making module selects the final coordinated operation plan based on the Pareto optimal solution set according to the preset preference.

10. The multi-objective collaborative optimization operation system of a gas generator set according to claim 9, characterized in that, It also includes a multi-source heterogeneous data fusion module and a dynamic scheduling execution unit. The multi-source heterogeneous data fusion module is used to integrate the position information and the output power; The dynamic scheduling execution unit is used to control the power output and start-stop timing sequence of the gas generator sets according to the final coordinated operation plan.

Images