مقایسه عملکرد مدل‌های DP، SDP و SSDP در بهینه‌سازی بهره‌برداری از مخزن‌های آبی چند‌منظوره (مطالعه موردی: مخزن سد زاینده‌رود)

نویسندگان

چکیده

مدل‌های برنامه‌ریزی پویا ابزاری مناسب برای تعیین سیاست‌های بهینه بهره‌برداری از مخزن‌های سدها،‏ به‌خصوص در مواجه با محدودیت و عدم قطعیت در منابع آبی هستند. در این مقاله،‏ عملکرد طیف متنوعی از این مدل‌ها شامل برنامه‌ریزی پویای قطعی (DP)‎،‏ برنامه‌ریزی پویای احتمالاتی با کلاسه‌بندی جریان (SDP_Class)‎،‏ برنامه‌ریزی پویای احتمالاتی مبتنی بر سناریوهای تاریخی جریان (SDP_Scenario)‎ و برنامه‌ریزی پویای احتمالاتی با نمونه‌گیری (SSDP)‎ در سیستم چند‌منظوره سد زاینده‌رود ارزیابی و مقایسه شده است. ابتدا سیاست‌های بهینه بهره‌برداری در شرایط نیاز ثابت کشاورزی و سپس ملاحظه همزمان نیازهای کشاورزی و برقابی در کنار تأمین نیازهای شرب،‏ صنعت و زیست‌محیطی،‏ درنظر گرفته شد و در نهایت عملکرد مدل‌های بهینه‌سازی با استفاده از مدل‌های شبیه‌سازی و معیارهای کارایی همانند اعتمادپذیری زمانی و کمی،‏ بیشینه و میانگین کمبود و نیز زمان اجرای مدل‌ها ارزیابی شد. نتایج برتری محسوس مدل‌های SDP و SSDP را در مقایسه با مدل‌های DP نشان داد. این برتری عملکرد برای مدل SSDP و نیز شرایط حدی کم‌آبی‌ها بارزتر بود. با وجود عملکرد بهتر مدل SSDP،‏ بار محاسباتی و زمان اجرای آن بیشتر بود که روش‌های پیش‌تخصیص‌دهی متغیرها و برنامه‌نویسی برداری شده،‏ توانست تأثیر زیادی در کاهش زمان اجرای تمامی مدل‌ها و به‌خصوص SSDP داشته باشد.

کلیدواژه‌ها


عنوان مقاله [English]

Comparing the performance of the DP, SDP and SSDP models for optimizing the operation of multi-purpose reservoirs (Case study: Zayandeh-Rood dam)

نویسندگان [English]

  • Saeid Morid
  • Sedigheh Anvari
  • S.Jamshid Mousavi
چکیده [English]

Dynamic Programing models are appropriate tools for deriving optimal operation policies of reservoirs, especially during the water scarcity situation. In this paper, the performances of a variety of these models including deterministic Dynamic Programming (DP), Stochastic Dynamic Programming with inflow classification (SDP_Class), Stochastic Dynamic Programming based on inflow scenarios (SDP_Scenario) and Sampling Stochastic Dynamic Programming (SSDP) were evaluated and compared for Zayandeh-Rood multi-purpose system. For this aim, beside the complete supply of domestic, industrial and environmental demands, optimal operation policies considering constant agricultural demand and then both agricultural and hydropower ones were derived. Finally the performance of optimization models was evaluated and compared based on simulation models, computational load of optimization models as well as some efficiency criterion such as temporal and volumetric reliability, maximum and average shortage. Results showed that stochastic models like SDP and SSDP performed better than DP models, considerably. This superiority was considerable, especially during extreme water shortage conditions. Despite the SSDP preference its computational load as well as its execution time were higher than others. In order to alleviate this problem, pre-allocation and vectorization techniques were applied and they had a significant effect on the run-time reduction of all models, particularly on SSDP.

Dynamic Programing models are appropriate tools for deriving optimal operation policies of reservoirs, especially during the water scarcity situation. In this paper, the performances of a variety of these models including deterministic Dynamic Programming (DP), Stochastic Dynamic Programming with inflow classification (SDP_Class), Stochastic Dynamic Programming based on inflow scenarios (SDP_Scenario) and Sampling Stochastic Dynamic Programming (SSDP) were evaluated and compared for Zayandeh-Rood multi-purpose system. For this aim, beside the complete supply of domestic, industrial and environmental demands, optimal operation policies considering constant agricultural demand and then both agricultural and hydropower ones were derived. Finally the performance of optimization models was evaluated and compared based on simulation models, computational load of optimization models as well as some efficiency criterion such as temporal and volumetric reliability, maximum and average shortage. Results showed that stochastic models like SDP and SSDP performed better than DP models, considerably. This superiority was considerable, especially during extreme water shortage conditions. Despite the SSDP preference its computational load as well as its execution time were higher than others. In order to alleviate this problem, pre-allocation and vectorization techniques were applied and they had a significant effect on the run-time reduction of all models, particularly on SSDP.

Dynamic Programing models are appropriate tools for deriving optimal operation policies of reservoirs, especially during the water scarcity situation. In this paper, the performances of a variety of these models including deterministic Dynamic Programming (DP), Stochastic Dynamic Programming with inflow classification (SDP_Class), Stochastic Dynamic Programming based on inflow scenarios (SDP_Scenario) and Sampling Stochastic Dynamic Programming (SSDP) were evaluated and compared for Zayandeh-Rood multi-purpose system. For this aim, beside the complete supply of domestic, industrial and environmental demands, optimal operation policies considering constant agricultural demand and then both agricultural and hydropower ones were derived. Finally the performance of optimization models was evaluated and compared based on simulation models, computational load of optimization models as well as some efficiency criterion such as temporal and volumetric reliability, maximum and average shortage. Results showed that stochastic models like SDP and SSDP performed better than DP models, considerably. This superiority was considerable, especially during extreme water shortage conditions. Despite the SSDP preference its computational load as well as its execution time were higher than others. In order to alleviate this problem, pre-allocation and vectorization techniques were applied and they had a significant effect on the run-time reduction of all models, particularly on SSDP.

Dynamic Programing models are appropriate tools for deriving optimal operation policies of reservoirs, especially during the water scarcity situation. In this paper, the performances of a variety of these models including deterministic Dynamic Programming (DP), Stochastic Dynamic Programming with inflow classification (SDP_Class), Stochastic Dynamic Programming based on inflow scenarios (SDP_Scenario) and Sampling Stochastic Dynamic Programming (SSDP) were evaluated and compared for Zayandeh-Rood multi-purpose system. For this aim, beside the complete supply of domestic, industrial and environmental demands, optimal operation policies considering constant agricultural demand and then both agricultural and hydropower ones were derived. Finally the performance of optimization models was evaluated and compared based on simulation models, computational load of optimization models as well as some efficiency criterion such as temporal and volumetric reliability, maximum and average shortage. Results showed that stochastic models like SDP and SSDP performed better than DP models, considerably. This superiority was considerable, especially during extreme water shortage conditions. Despite the SSDP preference its computational load as well as its execution time were higher than others. In order to alleviate this problem, pre-allocation and vectorization techniques were applied and they had a significant effect on the run-time reduction of all models, particularly on SSDP.

کلیدواژه‌ها [English]

  • Optimization-Vectorization-Pre-allocation-SDP-SSDP-DP-Zayandeh-Rood Dam.-