<p>Despite their non-renewability, high cost, and detrimental environmental impact, fossil fuels such as oil, gas, and coal are widely employed for energy-related purposes on a global scale. In contrast, despite being depletable, environmentally sustainable, and secure, renewable energy resources such as solar, wind, and hydro are still in their infancy on the energy market. With the potential for both small-scale and large-scale utilization, solar energy is the cleanest renewable energy resource available. For the sustainable utilization and extraction of solar energy, precise forecasting of solar irradiance is vital. As direct normal irradiance has a significant impact on solar power generation, it can be used to predict solar energy. However, the accurate prediction of direct normal irradiance is complicated by a number of factors, which makes it a difficult task. This investigation seeks to develop a novel hybrid model that combines the artificial bee colony optimizer and adaptive boosting algorithm in order to accurately predict direct normal irradiance. For the proposed model, input features were selected from ten features collected in Qinghai between July 1, 2022, and June 30, 2023. Comparing the alternative model in this investigation to the proposed model, which achieved the highest coefficient of determination value of 0.9790 during the testing phase, demonstrated its capability and efficacy. Thus, estimating direct normal irradiance for solar energy production can be accomplished with confidence using the proposed model.</p>