Research On Multi-objective Optimization Of Duration-Robustness For Underground Power Station Construction

The underground power station has the characteristics of crisscross structural layout,complex construction process and long construction duration.In the construction process,it’s faced with many unexpected interference events,such as construction machinery failure,rock burst,lining leakage,material damage and so on.The robustness of underground power station construction refers to the ability to resist the impact of emergency interference events and to be able to construct smoothly and orderly according to the schedule.Therefore,in the construction process of underground power station,how to ensure that the construction schedule has both strong robustness and short duration is a great challenge in the multi-objective optimization of construction schedule.The following key problems need to be solved:Firstly,how to determine the measurement criteria of duration and robustness;secondly,how to improve the schedule robustness when the construction schedule is being formulated,and how to realize the multi-objective optimization of duration robustness;thirdly,how to formulate the correction strategy to optimize the duration and robustness objectives in the process of construction schedule implementation.In response to the above problems,in-depth research on multi-objective optimization of duration-robustness for underground power station construction has been carried out and the following innovative research results have been achieved.（1）The composite robustness measurement method of underground power station construction schedule is proposed based on the improved prospect theory and Copula-CRITIC method.Firstly,to solve the problem that most of the current researches used a single solution robustness or quality robustness criterion to measure the robustness,a composite robustness criterion is proposed,which includes three criteria:starting time deviation r_s,structure deviation r_p and completion probability r_c.The comprehensive measurement of the schedule robustness is realized from two aspects:the solution robustness of the construction process and the quality robustness of the construction results.Secondly,interval theory is introduced into prospect theory to reflect the uncertainty caused by the incomplete cognition of experts.On this basis,a subjective weighting method of robustness criteria based on improved prospect theory is proposed;at the same time,the Copula-CRITIC method of objective weighting is proposed,which uses Copula function to replace the Pearson function in the CRITIC（criteria importance through inter criteria correlation）method,and overcomes the problem that it is unable to comprehensively consider the self strength and the correlation between r_s,r_p and r_c to carry out objective weighting;then,a fusion method of subjective and objective weights based on evidential reasoning which is improved by Jensen Shannon divergence and pan Zhou entropy is proposed.The difference between weights is measured by Jensen Shannon divergence,and the amount of information contained in subjective and objective weights is measured by Pan Zhou entropy.The reliability of each weight is comprehensively judged from two aspects of difference and amount of information,in order to improve the accuracy of weight fusion.The results show that:the number of optimized construction schedules with composite robustness criterion as the optimization objective is increased by 20,and the schedules have good robustness of solution and quality at the same time,which verifies the superiority of the composite robustness measurement method in this paper.（2）The buffer setting method of underground power station construction schedule under the framework of improved critical chain method is proposed.To solve the problem that the current buffer setting research has not set centralized buffer and decentralized buffer at the same time,and the existing buffer sizing methods lacks the scientific calculation of the probability of delay,failure and shutdown caused by interference events,this paper proposes the buffer setting method,by which setting project buffer PB at the end of the critical chain,setting feeding buffer FB at the intersection of non critical and critical chain,and setting scattered buffer SB before important activities on critical chain.Thus,the solution robustness and quality robustness from the construction process and results of underground power station can be improved.At the same time,Dezert Smarandache theory（DSm T）is used to solve the comprehensive probability of adverse consequences such as delay,failure and shutdown caused by interference events,which solves the problem of buffer size calculation based on experience of elastic coefficient method.Moreover,according to Fibonacci golden section theory,the total amount of buffers on critical chain calculated by DSm T-elastic coefficient method is allocated as the sum of decentralized buffers and project buffers;starting time criticality（STC）method is used to sort the criticality of activities on the critical chain,so as to determine the size and location of scattered buffers.The results show that:compared with the comparison method,the total amount of FB based on this method increases by 13.26days,the total amount of SB increases by 39.58 days,and the total amount of FB increases by 58.27 days;The robustness of the construction schedule is improved by27.71%,which verifies the superiority of the buffer setting method in this paper.（3）The hybrid grey wolf optimizer with sine cosine algorithm for duration-robustness of underground power station construction is proposed.To solve the problem that the existing research on the optimization of underground power station construction schedule lacks the common concern of the two mutual check and balance optimization objectives of duration and robustness,and the traditional grey wolf optimizer（GWO）is easy to fall into local optimum because of its slow convergence speed in the late iteration,the multi-objective optimization mathematical model of underground power station construction is established with the objectives of duration and robustness.Then,through the generation and switching of random number of sine function and cosine function,sine cosine algorithm is used to improve the first wolf^? in GWO,to balance the convergence and divergence ability of GWO and to overcome the problem of slow convergence speed and easily falling into local optimum in the late iteration.Finally,VIKOR method is used to select the best construction schedule from Pareto solution set.The results show that:compared with MOPSO,MOGWO,MOEA and MOMVO,HGWOSCA has shorter optimization speed（616.337 s）and more optimization solutions（34）,and has the advantages of solving the multi-objective optimization problem of construction period robustness of underground cavern group.（4）The multi-objective optimization method based on SD-DES schedule correction coupling model for duration-robustness of underground power station construction is proposed.To solve the problem that the existing research does not consider four kinds of corrective strategies,such as overtime,adding equipment,sending more personnel,and rescheduling;and the relevant system dynamics（SD）model fails to consider the construction personnel slack,equipment wears and maintenance and other correction parameters.Furthermore,only using SD model can not solve the problem of effective simulation and analysis of the process of"schedule deviation estimation-deviation correction strategy formulation-strategy implementation effect".Firstly,the SD model of corrective strategy is established,which includes four sub models:overtime,adding equipment,sending more personnel,and rescheduling.Secondly,in the overtime SD sub model,the correction parameters such as staff fatigue,slack and error are considered.Thirdly,the correction parameters such as wear,failure and maintenance are considered in the adding equipment SD sub model,and the bagging-MLP algorithm is proposed to predict the mechanical failure rate in the model.Finally,the SD-DES coupling model of schedule correction is constructed.Combined with the actual underground power station project,the application research of construction schedule correction is carried out.The simulation and analysis of"progress deviation estimation-deviation correction strategy formulation-strategy implementation effect"are realized,and the multi-objective optimization of duration-robustness of underground power station construction is realized.The results show that:compared with the traditional model,the accuracy of the rectification strategy based on this model is improved by 32.34%,which can guide the field rectification more effectively.